Skip to main content

Full text of "Hamadsha Vincent Crapanzano"

See other formats


THE 

HAMADSHA 

A Study in Moroccan 
Ethnopsychiatry 



Vincent Crapanzano 





THE HAMADSHA 
A Study in Moroccan Ethnopsychiatry 
By Vincent Crapanzano 

The Hamadsha, members of a Moroccan 
religious brotherhood that traces its spirit¬ 
ual ancestry back to two Muslim saints of 
the late seventeenth and early eighteenth 
centuries, have achieved a certain notoriety 
for their head-slashing and other acts of 
self-mutilation while in entranced states. 
Appealing largely to the illiterate Arab 
masses, the Hamadsha brotherhood has 
been regarded by scholars as a degenerate 
offshoot of the Islamic mystical tradition. 
Its beliefs and practices have been viewed 
as an unstructured amalgam of beliefs and 
practices of the ancient Mediterranean and 
of sub-Saharan Africa with others of Islam. 

Mr. Crapanzano sees Hamadsha beliefs 
and practices as forming a structured sym¬ 
bolic set interpretive of their experiential 
world. Unlike many other North African 
religious brotherhoods, the Hamadsha are 
concerned less with mystical union with 
Allah than with curing the demon-struck 
and the demon-possessed. Saints, demons, 
and a mana-like quality called baraka, "char¬ 
tered" by hagiographic legends and inte¬ 
grated with the socio-economic organization 
of the brotherhood, explain reactions that 
the Western observer would often classify 
as hysterical, depressive, or even schizo¬ 
phrenic. Saints, demons, and baraka pro¬ 
vide, however, more than an explanation of 
these reactions; they are the very "lan¬ 
guage" through which the reactions them¬ 
selves are immediately formulated and the 
cure effected. The cure itself—the dancing 
of the patient through a deep trance in 
which he becomes possessed and may even 
mutilate himself—serves to incorporate him 
into a cult, the Hamadsha brotherhood, 
which provides him not only with opportu¬ 
nities to discharge those tensions presum¬ 
ably responsible for his illness but also with 
a reinforced symbolic interpretation of his 
existence. 

VINCENT CRAPANZANO is a member 
of the Department of Anthropology at 
Princeton University. 


THE HAMADSHA 




THE 

HAMADSHA 

A Study in Moroccan 
Ethnopsychiatry 

VINCENT CRAPANZANO 


UNIVERSITY OF CALIFORNIA PRESS 
BERKELEY, LOS ANGELES, LONDON 



X0DP-25Z-2NCG 




UNIVERSITY OF CALIFORNIA PRESS 
BERKELEY AND LOS ANGELES, CALIFORNIA 
UNIVERSITY OF CALIFORNIA PRESS, LTD. 

LONDON, ENGLAND 

COPYRIGHT <£> 1973, BY 

THE REGENTS OF THE UNIVERSITY OF CALIFORNIA 

ISBN: 0-520-02241-6 

LIBRARY OF CONGRESS CATALOG CARD NUMBER! 72-75529 
PRINTED IN THE UNITED STATES OF AMERICA 



Contents 


Preface 


A Note on the Rendering of Arabic Words and Phrases 

Yll 

ix 

A Hamadsha Performance 

xi 

Introduction 

1 

Part One 

1. Historical Origins: Sufism 

15 

2. The Saints and the Orders: Their History 

_22 

3. The Legends 

30 

Part T*wn 

4. The Saints* Villages 

61 

5. The Lodges of Meknes 

75 

6. The Shantytown Teams 

101 

7. The Circle of Exchange 

114 

Part Three 

8. The Theory of Therapy 

133 

9. The Pilgrimage 

169 

10. The Hadra 

185 

11. The Explanation of Therapy 

212 

Appendix 

231 

Glossary 

237 

Bibliography 

241 

Index 

242 

























Preface 


This study is based primarily upon field work with the Hamadsha of 
Meknes and the Jebel Zerhoun in 1967 and 1968. My research was sup¬ 
ported by a grant from the National Institute of Mental Health (MH 
13776-01), a Quain Grant from the Institute of Intercultural Research, 
and an award from The Committee on Research in the Sciences and 
Humanities of Princeton University. 

For preliminary advice I should like to thank Professor Ernest Gell- 
ner of the London School of Economics; Professor Clifford Geertz, 
then of the University of Chicago; Professors Abdelqadar Khatibi and 
Abdclouahcd Radi of the University of Rabat; Drs. Lawrence Rosen, 
Stuart Schaar, and John Waterbury; Mr. Erich Alport of Oxford; Mr. 
Ahmed El Yacoubi of New York and Tangier; and a nameless Sicilian 
waiter at a Perpignan cafe, who first told me that the Hamadsha were 
still in existence. To Mr. David Hart of Almeria I am especially grate¬ 
ful not only for his preliminary suggestions but for his continual en¬ 
couragement and advice throughout my term of field work. To Dr. 
Benykhalef, Secretary General of the Moroccan Ministry of Health; 
Dr. J. J. Maupome, medicin-chef of the Hopital el Ghazi in Sal6; Dr. 
George Brown; and Mr. William Stott, then of the American Embassy 
in Rabat, I am indebted for facilitating my research. 

Although I assume full responsibility for the contents and theoretical 
premises of this study, I should like to thank Professors Conrad Arens- 
berg, Margaret Mead, Robert F. Murphy, and Abraham Rosman of 
Columbia University; Dr. Theodora Abel of the Postgraduate Center 
for Mental Health in New York; Dr. Nicholas S. Hopkins and Dr. Dale 
F. Eickelman of New York University; and Mr. Roy Mottahcdeh, 
then of Harvard, for their many suggestions. Professors Mead, Mur¬ 
phy, Rosman, and Hopkins have read and commented on the text. To 
Dr. George Devereux of the Ecole Pratique des Hautes Etudes, I am 
particularly grateful. Not only have his keen psychological insights 



THE HAMADSHA 


viii 

been of great value to me, but his many comparisons with the ancient 
Greek world have added perspective and depth to my thinking about 
the Hamadsha. 

It is of course impossible to thank all my Moroccan friends who not 
only offered me their hospitality but devoted themselves to my study 
with patience and understanding. I am especially grateful to the mizwar 
of Sidi Ahmed Dghughi and the muqaddims of the Hamadsha lodges 
and teams of Meknes; to Ahmed bel Louafi of Beni Ouarad; Hamadi 
ben Salah and Hadda and I.abid ben Mohammed of Meknes; and 
Moulay Abdeslem ben Moulay Mahajub, who treated me as both son 
and student. 

Finally I should like to thank my wife, Jane Kramer, and my field 
assistant, who has asked to remain anonymous. Without them this study 
could never have been made, and to them I dedicate it. 



A Note on The Rendering 
of Arabic Words and Phrases 


Arabic words and phrases in this study have been rendered in the sim¬ 
plest manner yet recognizable to the Arabic speaker. Diacritical marks 
have been avoided wherever possible. Only the c ain, the ha, and the in¬ 
ternal glottal stop (by an umlaut) have been used. Place and tribal 
names and the names of my informants have been written in the French 
fashion currently in use in Morocco. Other words have been written 
in a manner more in keeping with English phonetics or as they appear 
in the Merriam-Webster Third New International Dictionary. With 
the exception of a few plurals such as “jnun,” “foqra,” and “ghiyyata,” 
which occur frequently in the text, all other plurals of Arabic words 
are indicated by adding an “s” to the singular form. Italics are generally 
used only on the first occurrence of a foreign word in each of its con¬ 
texts. A glossary appears at the end of the study. 

The reader will be able to approximate Moroccan Arabic pronuncia¬ 
tion by pronouncing consonants as in English and vowels as in Italian. 
The g is always hard as in “go” or “geese.” The j is pronounced like 
the “s” in “pleasure.” The q is like an English “k” but much further 
back in the throat. The gh is pronounced like the uvular “r” in French. 
The kh is like the German “ch” in “Bach.” The sh is like the “sh” in 
“ship.” c and h have no English or European equivalents. The c is pro¬ 
nounced a little like “a” in father; h, like an “h” in a loud stage 
whisper. The glottal stop indicates a break between vowels as in “uh 
oh.” 

In the pronunciation of place names the reader has only to remember 
that an initial ou is pronounced like a “w” and a ch like an “sh.” The 
final e is not pronounced. 




And they cried aloud, and cut 
themselves after their manner 
with knives and lancets, till 
the blood gushed out upon them. 

(1 Kings 18, xxviii) 



A Hamadsha Performance 


The square in front of the tomb of Sheikh al-Kamal in Meknes was 
just beginning to fill up with townsmen, families from the nearby 
shantytowns, and a few Berber and Arab tribesmen when we arrived 
at 2:45 on a Friday afternoon in January. We were immediately sur¬ 
rounded by children—whom we had to fend off, sometimes violently, 
as they gaped and grabbed with curiosity at us. In one comer of the 
square a line of beggar women, huddled together, were blankly watch¬ 
ing a woman prance around to the wailing of four or five singers, 
hawking blessings for a few francs. Near them, a tiny, wizened old man, 
dressed in a white tunic, was neatly laying out a plastic tablecloth. He 
sat down on it, held up his staff between his legs, and wept. He was 
generally ignored. A circle of children had formed around another man 
who bandied a stick and shouted at them, and occasionally pulled open 
his shirt and puffed out his lungs through a round hole in his chest. 
Here and there crowds were starting to press around candy and orange 
vendors, con men and tricksters, story-tellers and preachers, dancers 
and fortune-tellers. In the comer closest to the tomb of their saint. 
Sheikh al-Kamal, the founder of the famed brotherhood of the e Isa- 
wiyya and the patron of Meknes, a group of adepts began their dance. 
As we moved through the crowd to watch them, we were suddenly 
attracted by the sound of the oboe known as the ghita and were told 
that the Hamadsha, whom we had come to see, were about to start their 
performance. 

We were greeted w r armly by Ali, a denizen of the nearby shanty¬ 
town whom we had met earlier in the day when he chased children 
from our car with a big stick. He shook our hands over and over again 
while the rest of the Hamadsha prepared their instruments. There w f ere 
nine in all: three guw'wala who played a large pottery drum shaped 
like an hourglass; one tabbal who played a snare drum; two ghiyyata 



xii 


THE HAMADSHA 


who played the oboe; two money collectors, Ali and another man who 
reminded me of a New England church usher; and a dance leader, or 
muqaddivi. The drummers were tightening their drums over a paper 
and cardboard fire as the ghiyyata tuned their instruments. Ali began 
to recite a prayer, or fatha , and to ask for money from the spectators 
who had gathered in a circle around the Hamadsha. He had a show¬ 
man’s sense of gesture and timing. 

Suddenly the ghiyyata began to play. It was now 3:11 p.m. The 
crowd of men, women, and children pressed inward and were vio¬ 
lently pushed back by Ali and the “usher” until a semi-circle was 
formed, with the musicians at one end, against the wall of the square, 
and a group of ten or eleven men standing shoulder to shoulder oppo¬ 
site them. The men raised themselves up on their toes and pounded 
down hard on their heels to the rhythm of the drums. At the same 
time, they raised and lowered their shoulders in a sort of ongoing shrug 
and hissed out air, occasionally chanting “Allah! Allah! Allah the 
eternal! Allah the adorable!” The muqaddim, a yellow-faced man 
dressed in a bright green acetate robe, danced directly in front of them, 
encouraging those who had fallen out of rhythm. Sometimes he would 
jump in the air, spin around, and land hard on his heels. At other times 
he would leap into the air and, as he landed, bring his outstretched 
fists in against his chest as though he were lancing himself. And at still 
other times he would pound his chest with his fists in a sort of breast¬ 
stroke motion. 

Almost immediately after the line of male dancers had formed, two 
women, one in a pale blue jallaba and the other in a black one, pushed 
their way through the crowd and began to dance directly in front of 
the ghiyyata. They did not move their feet as the men did, but instead 
bobbed up and down from the waist, their heads nearly hitting the 
ground, or swayed their bodies back and forth in much the motion that 
Arab women use to wash their floors. Their hair had come loose and 
was flying out in all directions. They reminded me of ancient maenads. 
Two other women joined them; all the women seemed to fall into 
trance much more quickly, and easily, than the men. 

By 3:30 there were four women dancing and the line of men had 
grown to 21. There must have been between 200 and 300 spectators 
standing in the circle and perched on the walls of the square. Ali and 
the “usher” made the rounds, collecting—almost extracting—a few 
francs from each of the spectators. The drumming remained constant, 
or so it seemed to me; it was the ghita which was producing the varia¬ 
tions in sound. The drumming, by this time, had begun to have a 



A HAMADSHA PERFORMANCE 


XUl 


dulling effect on me, and the music of the ghita an irritating one. I 
noticed that many of the spectators, especially those nearest the ghiy- 
yata, were in a light trance or at least dazed. Their eyes seemed glazed, 
fixed on the musicians or the dancers. The smell of all the hot, close, 
sweating bodies was stifling. 

The performance went on, without much variation, until a few min¬ 
utes after 4. Occasionally one of the male dancers would leave the line 
and dance in the center space, alone or with the muqaddim. Usually 
such dancers were in an entranced frenzy and were not able to follow 
the rhythm of the dance very well. One of the female dancers was led 
by a fat man, who participated only peripherally in the dance and 
seemed to be a sort of helper to the performers, over to the line of 
male dancers and made to dance with them. This seemed to relax her, 
to “bring her down.” 

At 4:15 there was a hush in the crowd as an extremely tall man in 
white robes, with a gold scarf around his neck, entered the dance area. 
A woman poked me and told me that he was a seer and a true Ham- 
dushi. A man signaled that he was a homosexual who played the pas¬ 
sive role. His costume was, in fact, effeminate, his breasts well-devel¬ 
oped, his hair long and curly, and his neck so swollen that I suspected 
some sort of glandular disorder. In a few minutes he was deep in a 
“chattering” trance: his mouth was opening and closing at a rate well 
out of the range of voluntary behavior. His head was thrust far back, 
his eyes were popping. He wandered, disoriented, around the center 
of the circle. Then the ghiyyata changed their tune slightly, and he 
was immediately “drawn” to them. He danced before them, his back 
to the audience, in a way which was closer to the women’s dance than 
to the men’s. He seemed more closed in upon himself than the other 
dancers, more separated from the audience and the other performers. 
Suddenly he began to beat his head with what appeared to be his fists 
but were in fact two pocket knives, one in each hand. The woman 
next to me whispered, “ ‘Ai’sha, c Aisha Qandisha.” Faster and faster he 
slashed at his head (the music too seemed faster), until his long curls 
were matted down with blood and his back and face were streaked 

with it. 

Many of the men and women looked on dispassionately, but the 
children in the audience grew restive and excited. More than one 
mother raised her baby high in her arms to see the slashing. The 
muqaddim began to dash frenetically around the perimeter of the circle. 
His eyes bulging, he asked for a knife, but the “helper” refused and, 
pulling the muqaddim toward him, took the leader’s head under his 



xiv 


THE HAMADSHA 


arm and scratched it. When the muqaddim finally regained his senses, 
the helper kissed him on the cheek and released him. By this time the 
head-slasher had stopped and was seated in a comer near some women, 
a very pained expression on his face. The musicians continued to play 
the same tune and in a minute or two he was up again, dancing and 
slashing with even more abandon than before. Then suddenly, unex¬ 
pectedly, he sat down again among the women. One of them began to 
bind his scalp with a pale blue scarf, another kissed his bloody hands 
and licked the blood that had stained her veil. A baby was lifted over 
the crowd and handed to the slasher, who kissed him. A third woman 
smeared a little blood on the baby’s stomach. The slasher no longer 
looked pained; his expression now was radiant. 

It was now 4:35. The musicians had changed their tune, and the 
dance seemed calmer to me. Twenty men were still pounding and 
hissing in their line. Several women had danced through the head¬ 
slashing scene, quite oblivious to it; one of them, a woman in black, had 
been bobbing up and down since the beginning of the performance, an 
hour and a half before. The rest of the dance seemed very unreal to me. 
I felt very distant, very removed from what was going on in front 
of me. 

At 4:55 the ghiyyata blew two or three long, wailing blasts, and the 
performance was over. A few of the performers shook hands while the 
crowd dispersed. Several women came up to the slasher to ask his 
blessing. The Isawa, the followers of Sheikh al-Kamal, were still danc¬ 
ing in their comer, but they had drawn a much smaller crowd than 
the Hamadsha. We were told that the Miliana, the followers of an 
Algerian saint who specialize in playing with and eating fire, had also 
performed, as well as a branch of the Isawiyya that charm snakes. 

Friday, January 12, 1968 
Sheikh al-Kamal, Meknes 
Morocco 



Introduction 


The Hamadsha are members of a loosely and diversely organized 
religious brotherhood, or confraternity, which traces its spiritual herit¬ 
age back to two Moroccan saints of the late seventeenth and early 
eighteenth centuries, Sidi £ Ali ben Hamdush and Sidi Ahmed Dghughi. 
Despite a certain notoriety due to their head-slashing and other prac¬ 
tices of self-mutilation, the Hamadsha have received comparatively 
little attention in the literature, ethnographic or other, on Morocco and 
North Africa. 1 This has probably resulted less from any secretiveness 
or lack of cooperation on their part than from their political insignifi¬ 
cance and from the fact that they have been overshadowed by larger, 
more spectacular brotherhoods like the ‘Isawiyya. 

The Hamadsha have been classified by French scholars as an extreme 
example of the confrerie populaire , a sort of degenerate form of the 
Sufi brotherhoods of the Muslim high tradition, corrupted by the base 
imagination of le peiiple, by survivals from the ancient religions of the 
circum-Mediterranean culture area, and by pagan influences from sub- 
Saharan Africa. They are considered, then, to be part of the cult of 
saints, or ?mraboutism, which has been generally regarded as the hall¬ 
mark of Maghrebian Islam. The French word “maraboutisme” is derived 
from the Arabic murabit, which describes a man attached to God—the 
root itself means “attach” or “fasten”—and has been used for any of 
the warrior-saints who brought Islam to Morocco. “Maraboutisme” has 
become in French a catch-all expression for all sorts of activities asso¬ 
ciated with the worship of saints. It may, for our purposes, serve to 
define two basic institutions: the cult of saints and the religious brother¬ 
hoods. 

The saints of Morocco—they are referred to as siyyid, salih , or wali 

1 The most complete study is a 19-page article by Herber, published in 1923. 



2 


THE HAMADSHA 


—may be descendents of the Prophet, founders and sheikhs of religious 
brotherhoods like the Hamdushiyya, 2 political heroes of the past, 
scholars reputed for their piety and religious learning, holy fools, or 
“simply vivid individuals who had tried to make something happen” 
(Geertz 1968:8). Associated with the tribal structure of Morocco, they 
run a gamut of importance from the purely local saint about whom all 
but his name is forgotten, and who is perhaps visited by half a dozen 
women each year, to a saint like Moulay Idriss, to whom all Moroccans, 
Berbers and Arabs alike, pay homage (Dermenghem 1954:11-25). 
Some, like Moulay Abdeslem ben Aleshish, Sidi Harazam, or Sidi Said 
Ahansal, were historical figures of considerable fame; while others, as 
Westermarck (1926 (I):49) put it, seem to have been invented to 
explain the holiness of a place. 

The object of the cult of saints is the saint’s tomb—usually a squat, 
white cubical building with domed roof ( qubba ). These dot the Mo¬ 
roccan countryside and are cared for by the saint’s descendants— 
celibacy is not a prerequisite for sainthood in the Islamic world—or by 
a caretaker ( muqaddim ) who lives on part of the alms received from 
pilgrims. The tombs are visited and venerated by men, women, and 
children anxious to obtain from their saint some favor such as a male 
child, a cure for a bout of rheumatism or a case of devil-possession, a 
favorable verdict at court, political asylum, or simply good fortune. 
A particular behavioral set designed to enable the pilgrim to obtain 
the saint’s blessing or holiness {baraka) is associated with each tomb. 
Its components may vary from the offering of a candle to the sacrifice 
of a bull or even a camel; from kissing the four sides of the tombstone 
to chanting long litanies; from rolling a holy stone over aching parts 
of the body to receiving massages from descendants of the saint. Sacred 
springs and grottos, trees, stones, and animals believed to contain 
baraka, and spots to which the jnun, or devils, are said to gravitate, 
are often found near the tombs. These too have their behavioral dic¬ 
tates which are linked to the veneration of the saint (Basset 1920). 

The brotherhoods are associated with the cult of saints, for their 
members follow the path ( tariqa ) of a spiritual leader, or sheikh , who 
is usually considered to be a saint. There is considerable variation in 
the organization, function, degree of theological sophistication, and 
ultimate aim of the brotherhoods. The members of the more sophis- 

3 The Hamadsha brotherhood is referred to as at-tariqa al-Hamdushiyya, or 
simply as l-Hamdushiyya. A male adept of the brotherhood is a Hcnndushi; a 
female adept, a Hamdushiyya. The plural for both male and female adepts is 
Harnadsha, which I also employ as an adjective. 



INTRODUCTION 


3 


ticated are recruited, as might be expected, from the wealthiest, best- 
educated strata of Islamic society; the members of others, like the 
Hamadsha, come from the illiterate masses. All of the orders involve 
certain ritualized acts: the mechanical recitation of supernumerary 
prayers, reminiscent of the Sinaitic and Anthonie prayers of Jesus or 
the chants of mantra yoga; listening to music; dancing. The popular 
orders tend to be extreme: wild dances inducing ecstatic, frenetic 
trances; drinking boiling water; eating spiny cactus and other defile¬ 
ments; charming poisonous snakes; and innumerable acts of self-mutila¬ 
tion. All of them attempt to produce some sort of extraordinary psy¬ 
chic state which may be interpreted as union with God or possession 
by a demon. 

Unlike the members of the more sophisticated orders, who consider 
their founding saint as a spiritual master who has provided them with 
a path to God, the members of the popular orders often consider their 
saint as an object of devotion in his own right and the source of power 
for their miraculous feats. Some of the orders have an extensive net¬ 
work of lodges located not only in Morocco but as far East as Mecca 
and deep into sub-Saharan Africa; others are limited to a few members 
who meet when and where they can. Some have close ties with the 
descendants of their founding saint, to whom they must give their 
complete allegiance and all of the alms they collect each year; others 
have almost no contact with the families of the saint. Some have a very 
elaborate hierarchy of initiates; others no hierarchy whatsoever. Some 
meet in well-constructed lodges, others in the open or in private houses. 
All of them are firmly convinced that they are faithful members of the 
Orthodox Muslim community. Some of these confraternities still flour¬ 
ish today, others are moribund, and still others defunct. 

The Hamadsha are, in fact, members of two distinct brotherhoods 
which are closely related to each other and often confused. The c Alla- 
liyyin are the followers of Sidi c Ali ben Hamdush, and the Dghughiyyin 
follow Sidi c Ali’s servant, or slave, Sidi Ahmed Dghughi. 3 Both saints 
are buried and venerated some 16 miles by road northwest of the city 
of Meknes on the south face of the Zerhoun massif—Sidi c Ali, whose 
tomb is one of the largest in Morocco, in the comparatively wealthy 
village of Beni Rachid, and Sidi Ahmed about a mile farther up the 
mountain, in the much poorer village of Beni Ouarad. The inhabitants 
of Beni Ouarad are much darker than those of Beni Rachid. Roughly 

* I shall use the term Hamadsha to refer to both orders and to practices com¬ 
mon to both. c Allaliyyin and Dghughiyyin will refer to the specific orders and 
their specific practices. 




4 


THE HAMADSHA 


a sixth of the population of each village claims agnatic descent from 
their respective saints. They are collectively referred to as the imilad 
siyyid, the children of the saint. As the “children” of both saints claim 
descent not only from their saintly ancestors but from the Prophet as 
well, they may also be called shurfa , the Moroccan Arabic plural for 
sharif , a descendent of the Prophet Muhammad through his daughter 
Fatima and his son-in-law c AIi. Since the children of each saint are all 
able to trace their descent back to a single ancestor, they constitute, in 
anthropological terminology, a maximal lineage. Each of the two maxi¬ 
mal lineages, which are in turn divided into a number of smaller patri- 
lineages, is governed by a headman, or mizwar. A descendant of Sidi 
'Ali or Sidi Ahmed has the option of becoming a member of his an¬ 
cestor’s brotherhood, but rarely takes this option. 

The members of the Hamadsha brotherhoods—they are most com¬ 
monly called foqra —are divided into teams. A team ( tdifa ) may have 
a specific meeting place, or lodge, called a zawiya. Although the word 
“zawiya” refers, strictly speaking, only to the meeting place of a par¬ 
ticular ta'ifa, I will follow common Moroccan usage and use it to refer 
to members of a particular lodge as well. The tai’fa must also be dis¬ 
tinguished from the tariqa, which is either a brotherhood or the “path” 
or “way”—that is, the teachings—of a particular saint. 

Although the Hamadsha may be related historically to the mystical 
tradition of Islam, they do not usually conceive of the goal of their 
practices as union or communion with God, but rather as the cure of 
the devil-struck and the devil-possessed. They are essentially curers, and 
it is in this spirit that I propose to examine them. This is not to say 
that the Hamadsha would consider such an investigation appropriate or 
even desirable. They have received their power ( baraka ) to cure from 
Allah by way of His servant, their saint and intermediary to Him, and 
they are content with their lot. The ways of Allah are not to be 
questioned. To ask whether they conceive of their cures as essentially 
religious in nature, however “religious” may be defined, is to ask a 
question which has no meaning for them. All activities are religious 
insofar as they are contingent upon the will of Allah, and this very 
contingency is brought home to them with particular poignancy by the 
fact that the cures they effect are extraordinary, outside the tone and 
content of everyday life. 

The Hamadsha are not just curers but successful curers at that, in 
terms of the standards their society sets and, in some instances, in terms 
of the standards set by modem medicine. They are able to effect, often 
dramatically, the remission of symptoms—paralysis, mutism, sudden 



INTRODUCTION 


5 


blindness, severe depressions, nervous palpitations, paraesthesias, and 
possession—which led the patient or his family initially to seek their 
help. The symptoms they treat are frequently expressions of the com¬ 
mon anxiety reaction found in many primitive societies (Wittkower 
1971) or expressions of more severe hysterical, depressive, and even 
schizophrenic reactions. The Hamadsha are, in their own fashion, su¬ 
perb diagnosticians and generally avoid treating those illnesses which 
are regarded by Western medicine as organically caused. They seldom 
treat epilepsy. 

The Hamadsha complex is to be regarded here, then, as a system of 
therapy. Therapy is considered to be a structured set of procedures for 
the rehabilitation of an incapacitated individual—an individual who is, 
from a sociological perspective, unable to meet role expectations and 
effectively perform valued tasks (Parsons 1964). Therapeutic pro¬ 
cedures effect changes in the ailing individual’s social situation as well 
as in his physical and psychological condition. He is moved through 
the roles of sick person and patient back, in the case of successful treat¬ 
ment, to his original role. If the treatment is not completely successful, 
he may be regarded as “a chronic case,” or as handicapped. The ideal 
is of course full restoration to his “old self.” 

Certain therapies, however, of which the Hamadsha is but one of 
many examples, may often be incapable of, or do not even aim at, re¬ 
storing the distressed individual to his previous condition. Rather, they 
introduce him to a new social role and concomitant tasks. The individ¬ 
ual may become a member of a cult like that of the Hamadsha. He is 
provided thereby not only with a new social identity but also with a 
new set of values and a new cognitive orientation—that is, with a new 
outlook. This new “outlook” may furnish him with a set of symbols 
by which—in the case of psychogenic disorders, at any rate—he can 
articulate and give expression to those particular psychic tensions which 
were at least in part responsible for his illness. This symbolic set is 
closely related to the cult’s explanation of illness and theory of therapy. 

Aside from techniques designed to alter the physical and psycho¬ 
logical condition of the patient and his social situation, a therapy must 
provide the distressed individual, the curer or curers, and other mem¬ 
bers of the society with an explanation of the illness and a theory of 
cure. 4 In the case of cure by incorporation into a cult, such explana¬ 
tions may be considered the ideology, or belief system, of the cult. 
Berger and Luckmann (1967:113) have written: 

4 In what follows I am indebted to Berger and Luckmann’s The Social Con¬ 
struction of Reality (1967). 



6 


THE HAMADSHA 


Since therapy must concern itself with deviation from the “official” defini¬ 
tions of reality, it must develop a conceptual machinery to account for such 
deviations and to maintain the realities challenged. This requires a body of 
knowledge that includes a theory of deviance, a diagnostic apparatus, and a 
conceptual system for the “cure of souls.” 


To the extent to which such explanations are commonly known, or at 
least known to the ailing individual, they tend to formulate the illness 
and furnish, thereby, a ground for therapeutic procedures. This is 
particularly true of psychogenic disorders. 

In therapies like that of the Hamadsha the elements of explanation 
consist, as we shall see, of symbols which represent both social and 
psychic realities for the ailing individual (and other members of his 
milieu). These elements—images, in Godfrey Lienhardt’s term—serve 
not only to articulate but to interpret the individual’s experience imme¬ 
diately, and must be at once congruent with both psychological needs 
and socio-cultural realities. They are not individual projections. They 
are givens in the world into which the individual is born and, as such, 
serve from the start to mold his reality and to realize themselves in his 
psychic life. They provide a schema for the interpretation of his expe¬ 
riences and make them congruent with the realities of his world and 
that of other members of his culture. Their locus, which may be sought 
within the recesses of the soul or without—in the world, say, of saints 
and demons—may reflect the characteristic stance of an individual 
within a particular cultural tradition to others within his world (Crapan- 
zano 1971). Such explanations—they may be called symbolic-interpre¬ 
tive—are characteristic of many so-called primitive therapies and cannot 
be divorced from the curing practices themselves. Therapy, in such 
cases, involves the manipulation of symbols not only to give expression 
to conflicts within the individual, but also to resolve them (Levi-Strauss 
1963a). 

It is suggested here that the Hamadsha effect their cures by incor¬ 
porating their patients into a cult which provides them with both a 
new role—one which is probably more in keeping with their individual 
needs—and an interpretation of their illness and its cure. This inter¬ 
pretation permits during the curing ceremonies the symbolic expression 
of incapacitating conflicts and the consequent discharge of tensions 
which may impede social behavior. This discharge of tensions is not 
merely an emotional outburst, which may be of little therapeutic im¬ 
port, but a highly structured process which involves the symbolic 
resolution of such tension-producing conflicts. The process of resolu- 




INTRODUCTION 


7 


tion serves not only to “resocialize the deviant into the objective reality 
of the symbolic universe of the society,” as Berger and Luckmann 
(1967:114) maintain, but to reestablish or reinforce his motivation. 

It must be emphasized that the practices and, to a lesser extent per¬ 
haps, the beliefs of the Hamadsha and the members of other similar 
brotherhoods are not characteristic of “Moroccans” in general. The 
Hamadsha complex is a fringe phenomenon, peripheral but by no 
means unrelated to the mainstream of the Moroccan socio-cultural 
tradition. Many Moroccans, especially Berbers and the educated Arabs, 
look askance at the practices of the Hamadsha; they consider them to 
be uncouth, unorthodox, disgusting even, and are often embarrassed 
when reference is made to them by foreigners. Still, it has been my 
impression that even among the better-educated—though perhaps not 
among the best-educated—-disapproval is tempered by a certain awe 
which results, if for no other reason, from the dramatic quality of the 
Hamadsha performance and the “spectacular” nature of the Hamadsha 
cure. These performances, and these cures, are after all the will of God. 

The Hamadsha, who are almost exclusively Arabs, consider them¬ 
selves to be members of the Orthodox (Sunni) Muslim community and 
follow—or, perhaps more accurately, believe they follow—the laws and 
traditions of that community which find their inspiration, if not their 
very source, in the Koran and in the Prophetic tradition ( hadith ). 
Indeed, they find the very ground not only of their religious belief and 
worldview but of their social organization in the Koran—or, again 
more accurately, in what they impute to the Koran. The Hamadsha 
are in this respect not dissimilar to the millions of illiterate or quasi¬ 
literate Muslims of North Africa and the rest of the Middle East. As 
heterodox as their beliefs and practices may be, they do recognize the 
fundamental importance of the “five pillars” of Islam—profession of 
faith, prayer, almsgiving, fasting, and the pilgrimage to Mecca—and 
attempt to lead their lives accordingly. 

Like Arabs throughout the world, the Hamadsha are patrilineal and 
patrilocal; that is, they trace their ancestry through the male line and 
live, if not under the same roof as their fathers, then in the same village 
or neighborhood. People in the newly created shantytowns, however, 
often live tens if not hundreds of miles from their fathers’ homes; but 
even they still hold patrilocality as an ideal which they have had to 
abrogate by force of circumstance. Although the nuclear family tends 
to be the basic residential unit in the shantytowns—and to a lesser 
extent in the city and country—the extended family is perhaps the 




8 


THE HAM ADS HA 


basic social unit. Extensions beyond the extended family do not play an 
important role in the shantytowns and among the Hamadsha of the 
old quarter of Meknes; they do of course play an important role for 
the descendants of the Hamadsha saints. Genealogies of the former are 
shallow, seldom exceeding five generations; those of the latter are very 
extended, theoretically all the way back to the Prophet himself by way 
of his daughter. Parallel-cousin marriage, considered the ideal marriage 
among Arabs, is rare in all the Hamadsha settings. 

The father, or grandfather, as head of household, is all-powerful. 
He has strong and direct jural control over his wife or wives, and his 
sons (and daughters) must remain subservient to him until the very day 
of his death. No hostility whatever can be expressed toward him in his 
presence—or, for that matter, in his absence. Sons who are already 
middle-aged, for example, will not smoke in front of their father. His 
rule is absolute, and to foreign observers often appears arbitrary and 
harsh. There is considerable rivalry between brothers—often directed 
against the dominant one, who is usually but not necessarily the eldest 
—and this rivalry receives its fullest expression in disputes over in¬ 
heritance. (According to the Koran, all sons inherit equally; daughters 
inherit a half of what their brothers receive.) Sons were traditionally 
economically dependent upon and responsible to their fathers, who 

provided them with the bride-price necessary for marriage. Although 
this economic dependence is breaking down with wage-work in the 
cities and in Europe, economic responsibility is not. Sons still send home 
to their fathers much of what they earn. 

Women are considered inferior to men. Fathers—and mothers too 
—desire sons and not daughters, in spite of the fact that they receive a 
bride price upon their daughters’ marriages. Women are considered 
weak, defenseless, treacherous, and untrustworthy. They must be con¬ 
stantly watched, locked up even, by their husbands or male kin, and 
must always remain submissive to the aggressive dominance of their 
menfolk. (Sons at a very early age will begin to demand such a sub¬ 
missive attitude from their mothers.) Women are considered sexually 
insatiable by Moroccan men, at least by those of the Hamadsha’s milieu. 
The virginity of an unmarried girl—a symbol of her family’s honor— 
must be preserved at all costs. Wives must be prevented from amorous 
adventures. Fear of adultery is rampant. Lone women are always fair 
game. Although women are veiled and sequestered whenever possible, 
housing conditions in the bidonvilles —shantytowns which have grown 
up on the outskirts of most Moroccan cities, in the years following the 



INTRODUCTION 


9 


arrival of the French—give at least the illusion that they have more 
freedom than do the women of either the old quarter or the Zerhoun. 
Polygamy is rare in all the Hamadsha settings. Often it is desired by a 
wife who wants help in household matters. Usually there is rivalry 
between co-wives for their husband’s favor. Older sons are often re¬ 
sentful of their fathers’ second wives, or their stepmothers, who plot, 
they claim—not without justification—for their own children to inherit 
at the sons’ expense. 

Men must demonstrate no overt emotional dependence upon women; 
they must show no signs of femininity. They must strive continually 
to live up to the ideal of male behavior: domination; extreme virility; 
great sensitivity to matters of honor, independence, and authority; not 
to mention, of course, adherence to the canons of Islam. These ideals 
are embodied, realistically or not, in their image of their fathers. Indeed, 
the Arab male of the Hamadsha’s milieu is caught in a dilemma be¬ 
tween the dependence, the submission, the obsequiousness, even the 
passivity that he must show for years, often for more than half his 
lifetime, toward his father, and the independence, the domination, the 
authority, the aggressiveness that he must demonstrate to his sons and 
womenfolk. From the conceptual point of view, he must be at once 
both male and female. It is this dilemma that receives symbolic expres¬ 
sion, as we shall see, both in the hagiographic legends and in the Ha¬ 
madsha cures themselves. 

A few words on the nature and organization of this study are in order 
here. It is an attempt, on the one hand, to present the not-altogether- 
taken-for-granted world of the Hamadsha and, on the other hand, to 
uncover and make explicit the structures and symbols of that world. 
Its final aim is to offer an explanation, albeit hypothetical and incom¬ 
plete, of how the Hamadsha effect their cures. Substantively, the study 
is necessarily biased. My interest in the Hamadsha as essentially curers 
not only influenced my own perception of them and the questions I 
asked them but also the manner in which I have chosen to present the 
collected material. Naturally I have tried whenever and wherever pos¬ 
sible to compensate for this bias, but it would be foolish to claim that 
I have overcome it. The Boasian ethnography must always remain in 
the realm of the ideal. I was fortunate enough, however, to have had a 
view of the Hamadsha which was not altogether alien to their own 
view of themselves. Certainly in the shantytowns, and to a lesser extent 
in the old quarter of Meknes, the Hamadsha considered themselves to 



10 


THE HAMADSHA 


be primarily curers and were proudest of all of this activity. Their 
devotees were most taken with their extraordinary cures; and I am 
certain that I too was captured by their enthusiasm. 

Fieldwork itself was conducted in the standard anthropological fash¬ 
ion, with perhaps more than an average amount of material collected 
by free-association, imaging, and fantasizing techniques. My field as¬ 
sistant, a Berber, not a member of the Hamadsha brotherhood and not 
a permanent resident of Meknes or its environs, was present at many 
of my interviews. He was very gifted, endowed with a fine ethno¬ 
graphic curiosity and imagination and with that rare unobtrusive qual¬ 
ity that makes for a superb ethnographer. He did not serve as an in¬ 
formant. 

This study is divided into three main parts. Part One is concerned 
with the Hamadsha’s past, both in historical and legendary terms. 
Chapter One treats the Sufi tradition from which the Hamadsha are in 
part derived and with which at least the most knowledgeable Hamadsha 
recognize an affinity. It is a background chapter for the nonspecialist, 
and makes no new contribution to the study of the Islamic mystical 
tradition. Chapter Two relates what little of the history of the Hamad¬ 
sha is known. Chapter Three is concerned with the hagiographic leg¬ 
ends of the order, legends which are accepted by the Hamadsha and 
other Moroccans of their background as historically true. Indeed, it is 
the historical facticity of the legendary events that “charters” the Ha¬ 
madsha worldview and ritual activity. The legends are regarded here 
as givens in their world and provide the Hamadsha with a set of sym¬ 
bols, or perhaps more accurately with a justification for a set of 
symbols, by which they organize and give expression to at least part of 
their life situations, the most notable being of course their ritual activ¬ 
ities. The analysis of the legends is carried out from a combined struc¬ 
tural and psychoanalytic perspective which reveals, it is hoped, not 
only underlying themes, perhaps indicative of tensions within Moroc¬ 
can society and personality, but also the possible symbolic significance 
of certain elements that recur in the Hamadsha’s therapeutic theory 
and receive symbolic enactment in their rituals. 

Part Two is concerned with the component institutions of the 
Hamadsha complex and their intricate interrelationships. Descriptions 
of the order in the saintly villages of Beni Rachid and Beni Ouarad, as 
well as of the teams in their urban and shantytown settings, are given 
in Chapters Four, Five, and Six. The diverse personnel of the complex 
are described sociologically, and their relationship to one another and 
to the saints they worship is examined in detail in Chapter Seven. The 




INTRODUCTION 


11 


logic of baraka, or blessing, already found to be of singular importance 
in the legendary material, is related to the social and economic organi¬ 
zation of the order. A digression, not properly speaking pan of the 
phenomenological orientation of this study, describes how the Hamad- 
sha serve to integrate newcomers to the city by providing them with 
an enlarged social field and more complex interpersonal relations. 

Part Three is devoted to Hamadsha therapy. Chapter Eight, on the 
theory of therapy, attempts to present—in a manner comparable to the 
presentation of Western therapeutic practices—the Hamadsha’s own 
explanation of their cures. It is again concerned with the givens of their 
world. Particular attention is paid to the way in which members of the 
order, and other Moroccans, relate to the jnun, or demons, and how 
the jnun themselves are related to the saints. Baraka is found to be the 
curative element par excellence, but is in itself insufficient to effect a 
cure. Chapters Nine and Ten are devoted to a description of the curing 
rituals—the pilgrimage and the trance dance—both from the perspec¬ 
tive of an outside observer and from that of the actors themselves. 
“Elements” which occur in both the legends and the theory of therapy 
are given symbolic enactment in the rituals themselves. 

The final chapter of the book, Chapter Eleven, attempts a synthesis 
of the diverse components of the Hamadsha complex. It is predicated 
on the fact that every therapeutic system functions at all levels of 
human existence: the physiological, the psychological, and the socio¬ 
logical. Apart from whatever effects the Hamadsha’s rituals have on 
the physiological condition of their patients and themselves, they pro¬ 
vide them with a symbolic set, historically justified and socially and 
ritually reinforced, which is integrated with their social organization 
and expressive perhaps of tensions inherent in at least that segment of 
Moroccan society from which the Hamadsha are recruited. This sym¬ 
bolic set, it is suggested, serves to articulate and give expression equally 
to the Hamadsha’s experience of their physical and social, if not their 
physiological, environment. It enables them to act out, albeit symboli¬ 
cally, the scars of their past, and may indeed be of therapeutic import. 
How exactly the structured symbolic set functions for the individual is 
relegated to a sequel to this work. 




PART ONE 




1 


Historical Origins: 
Sufism 


Islamic mysticism is known as Sufism, a word derived from the Arabic 
for wool, suf, and originally applied to certain ascetics who wore 
clothes of coarse wool as a sign of penitence and worldly renunciation. 
The aim of Sufism, like that of other mysticisms, has been to realize a 
union with the Ultimate Reality or Godhead (fana fi-l-haqq). This 
union has both a negative and a positive aspect: it involves, on the one 
hand, an escape from the bondage of the phenomenal self and an ab¬ 
sorption or annihilation into the Divine (fana), and on the other hand 
a continuance of real existence in the Divine (baqa) (Nicholson 1963: 
18). The Sufi’s quest has been to bridge the distance between himself 
and his God, and Sufism itself has generally been considered by Islamic 
scholars as a reaction to the “cold” and formalistic tenets of Orthodox 
Islam, which lays great emphasis on the absolute gulf between man and 
God (Evans-Pritchard 1949:1-2; Gibb 1961:135). 

Historically, the origins of Sufism have been related to such diverse 
movements as Christianity, Neo-Platonism, Gnosticism, Zoroastrianism, 
and Buddhism. Within the Islamic tradition itself, it has been consid¬ 
ered an outgrowth of the ascetic pietism of the first centuries of Islam 
which—by stressing both the fear of God and man’s responsibility to 
the moral ideal, and not simply to the mechanical observance of the law 
—encouraged introspection of the moral motive and promoted an 
awareness of the inner life (Rahman 1966:154; Massignon 1934). The 
first Sufis were most often quietists who pursued their quest for God 
in isolation and usually through ascetic discipline. With the develop¬ 
ment of Islamic law and theology, however, a split developed between 
those members of the Muslim religious intelligentsia, the hilama or 
canonists, whose emphasis was necessarily impersonal, and those—the 



16 


THE HAMADSHA 


Sufis, properly speaking—who looked for a more personal basis for 
their piety (Rahman 1966:155-156). The canonists were displeased 
with the Sufis’ emphasis on the searching of conscience, since the Ko¬ 
ranic law had only legislated for an external tribunal and had no 
weapon against religious hypocrisy. They accused the Sufis of hetero¬ 
doxy because the Sufis held that intention was more important than act, 
practical example ( sunna ) superior to the law ( fard ), and obedience 
better than observance (Massignon 1934:682). 

From the ninth century on, a number of Sufi thinkers attempted to 
integrate Sufism with Orthodoxy—a movement culminated in the 
twelfth century by al-Ghazzali—and, as they won converts from the 
more orthodox position, the simple piety and gospel of love of the 
earliest Sufis were transformed into an elaborate mystical doctrine of a 
spiritual journey toward God (Rahman 1966:162). The way or path 
to God, like the scala sancta of St. John Climacus, consisted of a num¬ 
ber of stages such as repentance, abstinence, poverty, and trust in God, 
through which the mystic had to pass in order to achieve his goal. 
These stages (singular form, rrmqama ), which constituted the ethical 
discipline of the Sufi and had to be mastered by his own effort, must 
be distinguished from those states ( afrwal; singular form, hal) —medi¬ 
tation, nearness to God, intimacy, certainty, and the like—which 
formed a psychological chain and over which the mystic had no con¬ 
trol. They descend from God. Once the mystic has passed through all 
the stages and has experienced as many of the states as God has been 
willing to allow him, he is raised to a higher plane of consciousness 
called Gnosis ( Ttia'rifa) and Truth ( haqiqa ), in which he realizes that 
knowledge, knower, and known are one (Nicholson 1963:28-29). 

The fact that by following the Sufi path a man could realize a 
coalescence with the Divine led to the elaboration of the doctrine of 
sainthood—a doctrine which was to permit the introduction of Gnostic 
and Neo-Platonic ideas like the notion of the Perfect Man of late an¬ 
tiquity (Van Grunebaum 1954:23) and of pagan beliefs and practices 
like ancestor worship. The Arabic word wali, or saint, derived from a 
root which means “nearness,” is used to refer to an individual, living or 
dead, whose holiness has brought him near to God and who has re¬ 
ceived, as a special favor from God, miraculous gifts (singular form, 
karama). Less emphasis has been placed on the exemplary quality of 
the saint’s life than on his ability to perform miracles—a sign of God’s 
favor. The saints were said to form an invisible hierarchy upon which 
the order of the world rested and to be headed by a Qutb, or Axis, 
the greatest Sufi of the age, around whom the universe rotated. Careful 



HISTORICAL ORIGINS*. SUFISM 


17 


attention was paid, however, to the difference between saints and 
prophets and between saintly miracles (singular form, karama ) and 
prophetic miracles ( mifjizat ); and so long as these distinctions were 
kept, the Orthodox had no qualms about incorporating the doctrine of 
sainthood, which was not expressly forbidden by the Koran. From the 
tenth century on there developed a derivative notion which was not 
within the spirit of Orthodoxy: the absolute and unquestioning sub¬ 
mission to a sheikh, or spiritual master, who often founded his own 
particular path, or tariqa. 

The development of these “paths” has been traced, on the one hand, 
to a class of preachers and story-tellers (singular form, qiissa) who en¬ 
larged upon Koranic stories, and on the other hand to informal gather¬ 
ings ( halaqa) for religious discussion and recitation of religious for¬ 
mulae ( dikr ). These particular preachers succeeded in introducing new 
ideas, especially of Shi’ite origin, which were vigorously condemned by 
the more orthodox Muslims. By the middle of the ninth century many 
of the important ideas of Sufism had been developed and were being 
publicly taught in Baghdad and other centers of Islam, with great pop¬ 
ular success. From the simple Koranic recitations of the eighth century, 
there developed a complex congregational ritual which was designed to 
send the participants into religious ecstasy ( ivajd ). Some of these ex¬ 
ercises consisted, first and by way of preparation, of communal recita¬ 
tions of religious writings ( hizb ) and then of prolonged repetition of 
shorter phrases (dikr), until the words no longer made an impression 
on the senses of the participant and nothing but the form of the Divine 
Name was left. Others involved playing and listening to musical instru¬ 
ments (sama c ) ; dancing (raqs or hadra)\ tearing clothes (tairmq)\ acts 
of self-mutilation like those of the Hamadsha; and the contemplation 
of a young man, an ephebus, believed somehow to represent or embody 
the Divine (Ritter 1955:491-501). The more extreme practices were 
considered heretical by the canonists, and the legality of music and 
dance has long been debated both within and without the Sufi move¬ 
ment (Mole 1963). 

During the twelfth and tliirteenth centuries the first organized 
tariqas —or dervish brotherhoods, as they are sometimes called—began 
to appear. Famous teachers gathered around them in lodges disciples 
who, once initiated into their teacher’s way, started branches in other 
lands. The founder of a brotherhood was usually venerated as a saint, 
and his place of burial, like that of other saints, became a site of pil¬ 
grimage. Not only was he provided with a genealogy of spiritual 
authority, a silsila, which led back to the Caliph c Ali, but often also 



18 


THE HAM ADS HA 


with a socio-biological one which traced his ancestry back to the 
Prophet and thereby gave him the right to be called a sharif. Although 
his sons—there were no rules of celibacy in most of the brotherhoods 
—often inherited his position as leader, this was not universally the case. 
The adepts themselves, who owed absolute allegiance to their master, 
were divided into two principal groups: the disciples, usually literate, 
who were involved in the religious activities of the lodge and in the 
collection of revenues; and the lay members, usually illiterate, who 
were nominally attached to the order and participated on certain stated 
occasions in the ceremonial (Gibb 1961:152). 

Some of the brotherhoods not only spread throughout the Islamic 
world but served also to introduce their particular brand of Islam into 
non-Islamic lands (Gibb and Bowen 1957:75; Abun-Nasr 1965). Al¬ 
though the hierarchical organization and the ties between main and 
subordinate lodges were often loose, the brotherhoods—particularly 
during times of stress such as the French invasions of North and West 
Africa in the nineteenth century (Abun-Nasr 1965:1)—did bring to¬ 
gether, at least temporarily, structurally disparate social groups. This 
unifying function must not be exaggerated, however, for the orders did 
not have the organization to knit their branches together permanently 
(Gibb and Bowen 1957:78). Appealing largely to artisans and members 
of the lower classes (Gibb and Bowen 1957:182), they tended not to 
ally themselves with the ruling classes, and often served as a bulwark 
against political despotism (Rahman 1966:182). Thus the continual hos¬ 
tility between the Sufis and the canonists, supporters of the central 
government, must be seen not only along theological lines but also 
within its political context. 

Sufism, which followed its own particular course in Morocco and 
helped to diffuse Islam through the country, was intimately linked with 
the tribal structure and political development of the area. Mahdi ibn 
Tumart, the founder of the great Almohad dynasty of the twelfth cen¬ 
tury and successor to the Almoravids who had played an important 
role in the spread of Islam in Morocco, was deeply influenced by 
Sufism. The most influential of the earliest of the Moroccan Sufis was 
perhaps Abu Medyan of Tlemcen, who died toward the end of the 
twelfth century. Not only had he traveled to the East, where he be¬ 
came a disciple of Moulay Abdelqader al-Jilani (1077-1127 a.d.), the 
founder of the first organized brotherhood, the Qadiriyya, but he had 
also studied the works of al-Ghazzali at Fez under c Ali ben Hirzihim 
(Sidi Harazam) (Michaux-Bellaire 1921:143; Rinn 1884:211 et seq.). 



HISTORICAL ORIGINS: SUFISM 


19 


Although Abu Mcdyan’s teachings emphasized quite simply the abso¬ 
lute concentration on God and the abnegation of the world—“Say 
‘God’ and abandon all that is material or pertains thereto, if thou de- 
sircst to attain the true end” (Gibb 1961:158)—he served, at least sym¬ 
bolically, to synthesize the teachings of the three most influential 
mystics in Moroccan Sufism: al-Jilani, al-Ghazzali, and their spiritual 
ancestor al-Junayd. 

The Sufi movement, already powerful in the twelfth century, devel¬ 
oped during the fourteenth and fifteenth centuries its own particular 
Moroccan cast, which has come to be called “maraboutism.” A mara¬ 
bout, identifiable with the Sufi saint, is a man bound to God, and in 
Morocco as elsewhere in the Islamic world such a man is believed to 
be possessed of a miraculous force or power called baraka , or blessing 
—which, as we shall see, is not only transmissible to his progeny but 
has in certain special circumstances a contagious quality. Clifford 
Geertz (1968:44) has attempted to capture its meaning in impression¬ 
istic and somewhat romantic terms. He writes: 

Literally, “baraka” means blessing, in the sense of divine favor. But spreading 
out from that nuclear meaning, specifying and delimiting it, it encloses a 
whole range of linked ideas: material prosperity, physical well-being, bodily 
satisfaction, completion, luck, plenitude, and, the aspect most stressed by 
Western writers anxious to force it into a pigeonhole with mana, magical 
power. In broadest terms, “baraka” is not, as it has so often been represented, 
a paraphysical force, a kind of spiritual electricity—a view which, though 
not entirely without basis, simplifies it beyond recognition. Like the notion 
of the exemplary center, it is a conception of the mode in which the divine 
reaches into the world. Implicit, uncriticized, and far from systematic, it 
too is a “doctrine.” 

More exactly, it is a mode of construing—emotionally, morally, intellec¬ 
tually—human experience, a cultural gloss on life. And though this is a 
vast and intricate problem, what this construction, this gloss, comes down 
to, so at least it seems to me, is the proposition (again, of course, wholly 
tacit) that the sacred appears most directly in the world as an endowment 
—a talent and a capacity, a special ability—of particular individuals. 
Rather than electricity, the best (but still not very good) analogue for 
“baraka” is personal presence, force of character, moral vividness. Mara¬ 
bouts have “baraka” in the way that men have strength, courage, dignity, 
skill, beauty, or intelligence. Like these, though it is not the same as these, 
more even of all of them put together, it is a gift which some men have in 
greater degree than others, and which a few, marabouts, have in superlative 
degree. The problem is to decide who (not only, as we shall see, among 




20 


THE HAMADSHA 


the living, but also among the dead) has it, how much, and how to benefit 
from it. 

And one might add, rather more concretely: how it manifests or is 
believed to manifest itself. Although baraka is indeed in certain in¬ 
stances a quality, not always as positive and beneficial by either Mo¬ 
roccan or Western standards as Geertz’ definition might lead us to 
believe, it has at times, in the popular imagination at any rate, a more 
independent existence—and, in either case, a logic of its own which, 
as we shall see, must be situationally understood. 

The religious brotherhoods, whose members played an important 
role in the diffusion of baraka, first appeared in Morocco in the thir¬ 
teenth century and have more or less retained the form of the tradi¬ 
tional Sufi confraternities. The Portuguese invasions in the fifteenth 
and sixteenth centuries served as a spur to their development and dif¬ 
fusion among the urban and rural masses; and from the collapse of the 
Merinid dynasty in the fifteenth century to the establishment of the 
present, or Alawite, dynasty in the seventeenth, the brotherhoods 
played a dominant role in Moroccan political history (LeToumeau 
1958; Terrasse 1930). The anarchical conditions immediately preceding 
the reign of Moulay Ismail and following his death may have provided 
an atmosphere conducive to the development of such popular brother¬ 
hoods as the Hamadsha. 

Most of the principal extant orders of Morocco, including the Ha¬ 
madsha—the Jilaliyya and the Tijanniyya are the outstanding excep¬ 
tions—are said to be derived from al-Jazuli (d. 1465 a.d.?), a disciple 
of al-Shadhili (d. 1258 a.d.), himself a disciple of a disciple of Abu 
Medyan. The teachings of al-Shadhili—he himself left no writing and 
founded no order—seem to have been orthodox and to have emphasized 
devotion to God. He discouraged monasticism and encouraged his fol¬ 
lowers to pursue their worldly professions. The five principal points of 
his teaching were the fear of God in secret and in open; adherence to 
Prophetic custom ( sunna) in word and deed; contempt for mankind 
in prosperity’' and adversity; resignation to the will of God in matters 
great and small; and recourse to God in joy and sorrow (Margoliouth 
1934:247). 

The fame of al-Jazuli, which overshadows his master’s in Morocco, 
is perhaps less the result of the man than the epoch. Jazuli, who at the 
time of his death was the head of a large network of lodges in the 
south of Morocco, was buried in the Valley of the Sous. It is claimed 
that his disciple al-Shayyaf carried his master’s corpse on his cam- 



HISTORICAL ORIGINS: SUFISM 


21 


paigns for twenty years and illuminated it each night with a candle 
the size of a man. Seventy-seven years later, at the height of a period 
of religious fervor and political consolidation, his body was exhumed 
and transported to Marrakech by one of the Saadian sultans as a rally¬ 
ing point for anti-Portuguese sentiment (Michaux-Bellaire 1921:148; 
1927:58 etseq.). 

The Isawiyya, the oldest extant order which traces its origin to al- 
Jazuli, was founded by M’hamed ben c Isa al-Mukhtari, the Perfect 
Sheikh (Sheikh al-Kamal), around 1500 a.d., and has probably influ¬ 
enced the practices, if not the doctrine, of the Hamadsha. The Hamad- 
sha themselves trace Sidi 'Ali’s spiritual ancestry to al-Jazuli by way 
of Bu'abid Sharqi, the patron of horsemen (d. circa 1600 a.d.), whose 
tomb is located in the town of Boujad near Beni Mellel (Schoen 1937). 
The author of the Salnvat al-Anfas, quoting from the Saluk at-Tariq al- 
Qariyya , reports that Sidi c Ali had received instruction from Sidi Mu¬ 
hammad al-Hafyan, who was in turn instructed by Bu'abid Sharqi. 
Bu'abid Sharqi had received instruction both from his father, Sidi Abu 
al-Qasim al-Zari al-Jabiri al-Ratsami, the pupil of al-Tebba, and from 
Sidi Abdallah ben Fasi, himself a pupil of al-Tebba through his teacher 
al-Ghezwani (Paquignon 1911:534). Sidi 'Ali himself is said to have 
taken a disciple, Sidi Ahmed Dghughi. Around these two men there 
developed two quasi-distinct orders, the history of which is, as we shall 
see, almost totally unknown. 



2 


The Saints 
and the Orders: 
Their History 


Little is known historically of the lives of either Sidi c Ali ben Hamdush 
or Sidi Ahmed Dghughi. Neither the saints nor any of their descendants 
or followers have left any writings, and although the “children of Sidi 
c Ali” have in their possession a hand-written history of their ancestor, 
recently prepared by a scholar from Fez, it appears to be little more 
than a genealogical justification for their claim to be descendants not 
only of Sidi c AIi but of Morocco’s national saint Moulay Idriss and 
the Prophet himself. They are thus able to carry with impunity the 
title of sharif. 

Sidi £ Ali is, according to the document, 1 descended from Moulay 
Abdeslem ben Meshish, 2 the teacher of al-Shadhili himself. Sidi £ Ali 
spent ten years in a corner of the Qarwiyyin University of Fez (near 
the entrance which faces the Bab Semmarin), praying all night and 
fasting all day. He kept to himself, never moving or speaking to anyone. 
He repeated the shahada , or profession of faith, 18,000 times a day. 
(Such supernumerary exercises occur in the lives of most Islamic 
saints.) Then he moved to the village of Beni Rachid, on the south 
face of the Jebel Zerhoun (Mount Zerhoun), where he was visited by 
countless pilgrims from all over Morocco. They were careful to keep 
a respectful distance and never to sit down near him or talk to him. 

1 The content of the document was translated and paraphrased for me by the 
son of one of the lineage heads of die wulad (“children of”) Sidi ‘Ali. 

* To be consistent I spell “mulay” with an “ou”— moulay —in both personal titles 
and place names. Thus Moulay Idriss the saint and Moulay Idriss the village are 
spelled the same way. 



THE SAINTS AND THE ORDERS: THEIR HISTORY 23 


He did, however, like to hear men talk of the Prophet; he would nod 
his head up and down, and afterward rise and perform the hadra> or 
ecstatic dance. Then when he no longer knew what he was doing, he 
would recite “Allah! Allah! Allah the eternal! Allah the adorable!” 
His head would spin, and he would begin to perform miracles. His fol¬ 
lowing increased, fie never turned from his path until his death in 
either 1131 or 1135 h.j. (1718/19 or 1722/23 a.d.). All the inhabitants 
of Meknes, Moulay Idriss, and the other villages of the Zerhoun came 
to Beni Rachid for his burial. During the period of mourning, his tomb 
was visited every day by mourners who founded lodges dedicated to 
him all over Morocco and even in Algeria and Tunisia. He is said to 
have been primarily influenced by Bu £ abid Sharqi and to have influ¬ 
enced, among others, Sidi Ahmed Dghughi. 

The author of the Salwat al-Anfas has written of Sidi e Ali in the 
following words: 

Of their number is Abu 1-Hasan e Ali ben Hamdush, buried on the mountain 
of the Zerhoun near Meknes. He is to be classed (May God have mercy 
upon him) among the sheikhs of the mystical tradition in which the ecstatic 
trance [hat] is powerful. He liked mystical reunions [mwm £ ], ceremonies 
having the same object [ hadra ], and panegyrics to the Prophet. He had a 
taste for instrumental music. . . . 

At certain times, becoming like a lion, he hit people with whatever fell into 
his hands, be it a stick, stones, some sort of vase, or other objects; no one 
could approach him then. He accomplished a number of feats, celebrated 
miracles. His followers and his companions have reported his ecstasies and 
his mystical seances; his very numerous companions spread throughout 
many lands; each year with pious ardor they gathered around him. He had 
zawiyas in all countries; and he fashioned a number of virtuous and benefi¬ 
cent men, all of whom were illuminated [ madjazib ] or at least had the 
reputation of being so. (Paquignon 1911:533-534) 

Of Sidi Ahmed’s life even less is known. Although it has been claimed 
that he came from the Beni Dghugh of the Dukkala tribe (Schoen 
1937:93), the present leader, or mizwar, of Sidi Ahmed assures me that 
his ancestor in fact came from the Beni Hsen tribe and left family and 
property there. I have also been told that he came from Mdari in the 
Sahara. 3 Sidi Ahmed is said to have had three wives—one from the Beni 
Hsen, a second from Fez, and a third from the village of Beni Ammar 
on the Jebel Zerhoun—and to have left children by each of these 

• I have not been able to locate Mdari. 




24 


THE HAMADSHA 


women. According to the descendents of Sidi c Ali, Sidi Ahmed was 
chosen by their ancestor from among the pilgrims to Beni Rachid to 
be his servant or even his slave; according to the descendants of Sidi 
Ahmed, their ancestor was chosen to be Sidi c Ali’s pupil and friend. 
When Sidi c Ali died, Sidi Ahmed is reported to have slashed his head 
in despair. He died a few years after Sidi ‘Ali and is buried in the 
village of Beni Ouarad, a mile up the mountain from Beni Rachid. His 
“children” are also in possession of a document which traces their an¬ 
cestry back to the Prophet. Rumor has it that they paid for this 
genealogy with the few houses that Sidi Ahmed left them in Fez. 

Not only is little known historically about either the descendants or 
the followers of Sidi c Ali and Sidi Ahmed, but there is a curious absence 
of legendary material about them—which contrasts, as we shall see, 
with the rich legendary biography of the saints themselves. 

Sidi c Ali is said to have died celibate and childless—a fact which “his 
children” are anxious to disguise and which accounts in part for their 
contracting the scholar from Fez to do up their genealogy. Although 
the scholar is said to have found a marriage certificate for Sidi c Ali in 
the library of the shurfa of Ouazzane, it is commonly accepted, even 
among Sidi 'Ali’s “children” themselves, that Sidi c Ali died childless 
and that those who claim him as their ancestor are in fact descended 
from one of his brothers. 4 Some of Sidi Ahmed’s children are settled in 
Beni Ouarad. Those who remained among the Beni Hsen have lost all 
but nominal contact with those who settled in Beni Ouarad. 

I have been able to learn nothing of the origin and spread of the 
Hamadsha. The legends suggest only that Sidi c Ali founded zawiyas on 
his way from Bu c abid Sharqi in Marrakech to Beni Rachid, 5 and that 
pilgrims and mourners who visited Sidi 'Ali’s tomb established lodges 
throughout Morocco. No mention whatsoever is made in the legends of 
the origin and spread of the brotherhoods dedicated to Sidi Ahmed 
Dghughi. Montet (1902:12-13) lists the following brotherhoods which, 
he claims, are related to the Hamadsha: Sidiqiyyin , the followers of 
Sidi Muhammad es-Sadiq who came from the South and whose fol¬ 
lowers hit their heads against each other during their ceremonies; 
Riahin, the followers of Sidi l- c Amar Riahi of Meknes, whose followers 
stick knives and forks in their stomachs without blood flowing; and the 
Miliana, who follow Moulay Miliana. Quedenfeldt (1886:689) adds the 
t Alamin i the followers of Sidi Qadur al- c Alami of Meknes; the Sejini , 

*To be sure, Sidi c Ali’s children prefer to keep their doubts about their saintly 
ancestry secret. 

1 See Chapter Three. 



THE SAINTS AND THE ORDERS: THEIR HISTORY 25 


the followers of Sidi Hamid es-Sejini, also of Mcknes; and the Qasmin , 
the followers of Sidi Qasim bu ‘Asria. Neither author lists his criteria 
for affiliation, and none of the Hamadsha I talked to recognized any 
affiliation with these orders, many of which are by now defunct. The 
Miliana do not themselves acknowledge any ties with either Sidi e Ali 
or Sidi Ahmed, and the ‘Alamin, “a bourgeois order,” would be loath 
to admit any such ties. 

It seems reasonable to assume that the two saints of the Zerhoun 
consolidated practices of an ancient origin under a single banner and 
integrated them into the particular brand of Islam characteristic of the 
IViaghreb. A legend recorded by Michaux-Bellaire and George Salmon 
(1906:336) is suggestive of how a local cult was integrated with the 
larger cult to Sidi c Ali. Sidi c Ali Sanhaji, a disciple of Sidi e Ali ben Ham- 
dush who is buried in the valley of Lekhous near the city of Larache, 
was said to be the sheikh of Sidi Abderrahman al-Majdub, the putative 
ancestor of the Wulad Majdub—all of whom were, in 1906, Hamadsha. 

Sidi e Ali Sanhaji was a bandit who had already killed 99 people. At this 
time another man was chasing after a young girl who refused him. He swore 
that he would have her dead or alive. The girl died and was buried. Sidi 
e Ali Sanhaji was waiting around near the girl’s tomb, to perform some mis¬ 
deed. He saw the man approach the girl’s tomb in the night and immedi¬ 
ately hid himself and watched the man open the tomb, take out the virgin, 
tear off her shroud, and try to possess her. By a miracle which Sidi e Ali 
could not account for, the virgin placed her arm between the rapist and 
herself. The rapist cut off her arm. The virgin resisted with the other arm, 
but this too was cut off. She closed her legs, but the rapist cut these off too. 
e Ali became indignant and called upon God, saying that he could not find a 
better occasion to kill his hundredth victim, who would be his last and with 
which he would buy back the 99 others. He heard a voice which cried out 
three times: “Acquit yourself, O Father of Acquittals.” (It is from this that 
Sidi e Ali Sanhaji has come to be known as the Father of Acquittals.) e Ali 
then killed his hundredth victim, buried the virgin again, after having 
joined together her limbs which had been cut off, and replaced her shroud. 
Having covered her with earth, he planted an olive branch at the head of 
the tomb and said: “God Almighty, if tomorrow I find the branch green, 
I shall know that you have accepted my vow and pardoned me.” 

The following morning, e Ali found his stick transformed into an olive 
tree covered with leaves. He fell to the ground and paid homage to God. 
He got up enlightened and wandered among the tribes, preaching the 
divine word. He stopped at Sehulijin, where the inhabitants built him a 
zawiya which still exists. He died and was buried there. They built him a 
qubba (mausoleum) which fell a short time later, and when a second qubba 



26 


THE HAMADSHA 


was built and collapsed, Sidi e Ali Sanhaji appeared to the muqaddim and 
ordained that he should be without a qubba. (Michaux-Bellaire and Salmon 
1906:336) 


The muqaddim of the zawiya at Sehulijin was also the muqaddim of 
the Hamadsha, and the zawiya itself was one of the meeting places of 
the Hamadsha who were going to attend the annual pilgrimage to Sidi 
£ Ali at Beni Rachid. A second Hamadsha zawiya in the area had been 
the home of one of Sidi c Ali’s descendants who had settled there for a 
short time. In fact, many of the Hamadsha lodges are reputed to have 
been the home or burial place of one or more of Sidi £ Ali’s or Sidi 
Ahmed’s children, and are therefore the receptacles of some of the 
saint’s baraka. 

It is impossible to determine how widespread membership in the 
Hamadsha brotherhood was before the French arrival in Morocco in 
1912. Reports of early French observers indicate teams in Fez (Aubin 
1904:319; LeToumeau 1949:366), Rabat (Mercier 1906:122), Sale 
(Mercier 1906:135), Tangier (Salmon 1904:105), Casablanca (“Casa¬ 
blanca et les Chaouia” 1915:63), Moulay Idriss (Canal 1902:136), and 
in the Gharb (Michaux-Bellaire and Salmon 1906:334 et seq.). Captain 
Garcia Figueras (n.d.) maintains that, along with the £ Isawa, the Ha¬ 
madsha were the most important brotherhood in the western part of 
the Spanish zone. 

The presence in rural areas of the Hamadsha, as well as other popular 
brotherhoods, should be emphasized, since it is sometimes assumed that 
such orders are restricted to urban and shantytown settings. The fact 
that all the members of the Wulad Majdub, mentioned in the legend 
of Sidi £ Ali Sanhaji, were Hamadsha suggests that membership in 
tribal areas often followed tribal lines and may at times have been 
restricted to saintly lineages. I have heard that the Wulad Khalifa of 
the Gharb, who like the Wulad Majdub claim descent from a saint, 
today consider themselves to be Hamadsha, whereas at the time of 
Michaux-Bellaire’s investigations in 1913 they appear to have been pri¬ 
marily members of another popular brotherhood, the Jilaliyya (pp. 282 
et seq.). They were famous then, as they are today, for their complete 
abandonment in acts of self-mutilation during their ecstasies. Michaux- 
Bellaire, generally a keen observer, also noted that many of the Hamad¬ 
sha were black (p. 236). There is no mention of the Hamadsha among 
the Berbers. The early reports also indicate that the Hamadsha of the 
cities were then, as now, recruited primarily from the lowest strata of 
urban society (Salmon 1904:101; LeToumeau 1949:366; Aubin 1904: 



THE SAINTS AND THE ORDERS: THEIR HISTORY 27 


318-319). The town of Moulay Idriss appears to have been an excep¬ 
tion; there the Hamadsha came until recently from all strata of society 
(ben Talha 1965:22). In Fez they appear to have been street porters, 
blacksmiths, oven attendants, shoemakers, and tanners, and to have been 
associated with their respective guilds (LeToumeau 1949:366). 

Hamadsha practices as they are described in these early reports 
closely resemble what takes place today in Meknes. There appear, how¬ 
ever, to have been public processions in the cities, during which the 
Hamadsha would work themselves into frenzies and slash at their heads 
with single- and double-bladed axes, clubs, iron balls, and other objects. 
These processions, which usually took place before the annual pil¬ 
grimage, or musem , to the Zerhoun, were often combined with the 
processions of the Isawiyya, whose musem took place on the Prophet’s 
birthday, a week before the Hamadsha musem. I have been told that in 
Marrakech at the time of the Isawa musem, the Hamadsha followed the 
Isawa, and then a week later during the musem of the Hamadsha the 
Isawa followed. In Sale the Hamadsha would parade through the streets 
on the day preceding the musem on the Zerhoun, collecting money and 
candles which were sold at the end of the day. The proceeds were given 
to the descendants of Sidi e Ali. At sunset, the educated Hamadsha— 
there were about a hundred of these—would go to the zawiya to recite 
litanies. The uneducated would return home. The following day the 
Hamadsha from Rabat, across the river, would arrive in Said at the 
time of afternoon prayers and would perform the hadra there until 
evening prayers, when they were served a meal with which the cere¬ 
mony would end (Mercier 1906:135-136). These processions may have 
reinforced ties between brotherhoods or teams which were located in 
neighboring cities or quarters and between which there was undoubt¬ 
edly an element of competition 6 (ben Talha 1965:22). 

There are no figures for membership in the Hamadsha brotherhoods 
at the beginning of the Protectorate. Mercier (1906:122) cites figures 
for Rabat (200 adepts) and Sale (350 adepts, which he breaks into 100 
foqra and 250 kbuddam ) 7 (1906:135), but these figures should not be 

* See Part Two. 

r It is not altogether clear how Mercier distinguishes these two groups. He may 
be referring to the distinction between foqra (adepts) and muhibbm (devotees) 
(see Chapter Five). Or perhaps the Hamadsha of the time distinguished between a 
core group of at least semi-literate adepts and a peripheral group of illiterate 
members who were not familiar with the litanies and other esoteric matters. Such 
distinctions are not uncommon in the Algerian brotherhoods (Depont and Cop- 
polani 1897). 



28 


THE HAMADSHA 


taken too seriously—for, as we will see, membership is impossible to 
determine without defining exactly what we mean by it. Mercier’s 
figures suggest, nevertheless, that there were many more Hamadsha in 
Rabat-Sale in 1906 than there are today. 

The same criticism is applicable to George Draque’s report in 1938 
(p. 122). The Hamadsha appear then to have been the ninth largest 
brotherhood in Morocco, numbering about 3400. This figure is broken 
down for region: 


Oujda 

none 

Taza 

none 

Fez 

822 

Meknes 

1161 

Port Lyauty (Kenitra) 

552 

Rabat 

221 

Casablanca 

134 

Mazagan (Mohammedia) 

136 

Safi 

192 

Central Atlas 

27 

Marrakech 

120 

Tafilelt 

24 

Confins Algero-Marocains 

10 


These figures, which presumably refer to both the 'Allaliyyin and the 
Dghughiyyin, are undoubtedly low; they do give, however, some idea 
of the distribution of the Hamadsha in Morocco, and support the opin¬ 
ion of other observers and of the Hamadsha themselves. 

Although I have made no attempt to number the Hamadsha in Mo¬ 
rocco today, I was provided with lists of about 40 teams from the 
leaders of both brotherhoods. 8 This number should not be taken too 
seriously either. The leaders were reluctant to admit that some of the 
teams were defunct and that Dghughiyyin and 'Allaliyyin were com¬ 
bined in others, such as in the Marrakech lodges. Moreover, they did not 
list “renegade” teams which did not show them great allegiance, like 

"According to the list provided for the 'Allaliyyin, there were teams in Asila, 
Casablanca, Ceuta, Demnat, El Jadida, Essaouira, Fez, Ksar el Kebir, Larache, 
Marrakech, Meknes, Aloulay Bouchaib, Rabat, Safi, Sale, Tamazd (Ait Baha), 
Tangier, Taroudant, and Tetuan. Marrakech was said to have 2 teams and Fez 3. 
The list for the Dghughiyyin included Casablanca, Fez, Ksar el Kebir, Marrakech, 
Meknes, Moulay Idriss, Moulay Yacoub, Ouazzane, Rabat, Sale, Tangier, and 
Tetuan. Marrakech, Moulay Idriss, and Moulay Yacoub each had 2 teams, and 
Fez 3. 



THE SAINTS AND THE ORDERS: THEIR HISTORY 29 


the seven teams operative in the shantytowns of Meknes and others in 
the Gharb, or Western Plains. 

On the basis of the figures I obtained for Meknes and from casual 
observations of the Hamadsha in other parts of Morocco, I am able to 
state that membership in the Hamadsha brotherhoods is today smaller 
than it was at the turn of the century, especially in the cities of Fez, 
Rabat, Sale, and Tangier, and that the Hamadsha are most active in 
Meknes, the Gharb—especially in the area of Sidi Slimane, Dar- bel- 
Amri, and Sidi Yahya du Rharb—and on the Jcbel Zerhoun. The 
c Allahyyin are more numerous than the Dghughiyyin. The Hamadsha 
appear to be less active in the old quarters of the cities than in the 
shantytowns. This pattern generally holds true for other “popular” 
orders such as the Isawiyya, the Jilala, and the Miliana. One must be 
careful, however, not to assume that membership is increasing in the 
shantytowns. I have found that the majority of Hamadsha in the 
Meknes bidonvilles, for example, were Hamadsha before they came to 
the city or were at least familiar with their practices, and that there are 
comparatively few recruits among the second-generation inhabitants of 
the shantytowns. Since the oldest of these slums is only about 40 years 
old, no definite patterns can yet be discerned. 



3 


The Legends 


Although the lives of both Sidi e Ali and Sidi Ahmed are poor in histori¬ 
cal fact, they are, like the lives of most Moroccan saints of any im¬ 
portance, rich in legend. The legends themselves are rooted in a his¬ 
torical reality which has been greatly elaborated and distorted by the 
addition of popular themes. These themes, or traits flottants , as the 
Bollandist Hippolyte Delehaye has called them, appear to be limited in 
number and to recur again and again in the lives of not only the Moroc¬ 
can saints but of saints throughout the Islamic world (Voinot 1948; 
Goldziher 1880). Like the Christian saints and martyrs, the Islamic 
saints appear to have lost their individual character in their legendary 
form. Referring to Christian saints, Delehaye (1955:23) writes: 

Ainsi depouilles de leur individuality, isoles en quelque sorte du temps et de 
l’espace, enleves a leur cadre naturel, les personnages historiques prennent 
dans l’esprit du peuple une forme irreelle et sans consistence. Au portrait 
vivant et nettement caracterise que nous a legue l’histoire se substitue un 
etre ideal qui n’est que la personnification d’une abstraction; au lieu de 
l’individu, le multitude ne connait que le type. 

If, as Delehaye (1955:2) maintains, one of the hallmarks of the Chris¬ 
tian hagiographic legend is the goal of edification, we must be careful 
not to attribute the same goal to the Moroccan legends. The saint, as 
we have noted, is not considered a wali by virtue of his exemplary life 
but by virtue of the miracles he has performed. His life, in legendary 
form, is in fact often little more than a list of such miracles, which are 
considered to be proof of his saintliness. Burhan is often used by the 
uneducated Moroccan as a synonym for miracle ( karama ), but it usu¬ 
ally means “manifest evidence,” “decisive and irrefutable proof,” or 
“clear demonstration.” It also carries with it the Koranic connotation 



THE LEGENDS 


31 


of “shining light” or “brilliant manifestation” (Gardet 1960:1326). The 
legends create wonder and awe in the individual and affirm the position 
of the saint, whose role is of central importance in the lives of his 
devotees. The saint, an inexhaustible font of baraka, serves not only to 
effect cures, to eliminate barrenness, and to control the jnun and other 
spirits, but also as a reference point by which much of reality, especially 
in its spatial and temporal dimensions, is organized. The annual pil¬ 
grimages—and individual ones as well—mark the flow of time; direc¬ 
tions are often given by reference to saints’ tombs (cf. Eliade 1965: 
21-98). 

The legendary versions of a single saint’s life vary from region to 
region, from individual to individual, from circumstance to circum¬ 
stance. There are, to be sure, certain constant themes which recur again 
and again; the very consistency of these themes often appears to be 
less a product of the historical matrix upon which the legend is built 
than upon a theme’s reflection of social and psychological tensions. It 
is important, however, not to overemphasize a social and psychological 
analysis of the content and structure of the legends themselves at the 
expense of their rhetorical and demonstrative functions. Their reflec¬ 
tion, both in structure and content, of the Moroccan (and the individ¬ 
ual’s) social and psychological background only strengthens their per¬ 
suasiveness. 

The legends are told over and over again in marketplaces, among 
friends and members of religious orders, and at the saints’ tombs and 
lodges. Most often, to demonstrate or emphasize a particular point, only 
episodes of the lives are told. Throughout my field notes there are 
fragments from the lives of Sidi c Ali and Sidi Ahmed which were in¬ 
troduced into a conversation to convince me of one or another truth: 
the greatness of one of the saints, the superiority of Sidi £ Ali over Sidi 
Ahmed, Sidi £ Ali’s dependency on Sidi Ahmed, the curative power of 
one of the saints, or his special control over the jnun. These examples 
often reflected tensions that were at play within the society at large or 
within the individual himself and which were corroborated in the 
course of the interview. They appear to have served a symbolic-inter¬ 
pretative function for the narrator. 

The Legendary Cycle 

I have chosen to present here the most complete legendary life of 
the two saints that I obtained. It was told to me by the then recently- 
appointed muqaddim of the zawiya of Sidi c Ali in Meknes, Sidi Mo- 



32 


THE HAMADSHA 


hammed Touijer, who had developed a very strong sense of respon¬ 
sibility toward my work and was anxious to make a good impression. 
It is therefore more ordered and complete than any other legend that I 
have collected. Told, as it is, from the point of view of a follower of 
Sidi e Ali, it tends to play up the importance of Sidi e Ali with respect 
to Sidi Ahmed—a reflection of the social and economic competition 
between the descendants and followers of the two saints. I have in¬ 
cluded in notes which follow the legend both Muqaddim Touijer’s and 
my own explication of obscure pans of the legend, as well as imponant 
variations I have recorded at other times. 

Sidi c Ali was caid to Moulay Ismail Moulay Ismail was a ferocious 
man who cut off the heads of liars. One day he ordered Sidi c Ali to 
fetch Btfabid Sharqi. Sidi c Ali went to Sharqi with his troops. Sharqi 
asked Sidi c Ali if he had come to get him. Sidi c Ali answered that he had. 
Sharqi then gave Sidi c Ali a hand-mill and told him to turn it. Sidi c Ali 
turned the mill, and when he stopped, the mill continued to turn on 
its own. Sidi c Ali threw some grain into the mill. A black woman came 
up to him and exclaimed that the mill was grinding the grain by itself. 
Sidi c Ali told the woman to shut up and cursed her blind. The woman 
immediately became blind. When Sharqi saw that the woman was in 
fact blind, he threw Sidi *Ali out and immediately fell ill and vomited 
into a pail. Then he ordered the blind woman to get rid of his vomit. 
Feeling her way along the wall, she took the pail to the door, where¬ 
upon Sidi < Ali, who had been hiding there, grabbed the pail and drank 
all of the vomit. This is how he got the remainder of his baraka from 
Bu'abid Sharqi. 

Then Sidi c Ali began a long trip and arrived at the c Ayn Kabir [the 
Great Spring] and sat down. 

Someone came up to Sidi c Ali. His name was Sidi Ahmed Dghughi . 
Sidi c Ali told him to heat some water so that he could wash for his 
prayers. Sidi Ahmed asked if Allah would give him baraka if he did 
this. Sidi 'Ali assured him that He would. 

At this time the son of Moulay Ismail, c Abdelhaqq, was in the habit 
of shooting arrows into passersby. Before killhig them, he would invite 
them in to drink some buttermilk. Then he would shoot them. Sidi *Ali 
told Sidi Ahmed to go to the palace to do something about this. Sidi 
Ahmed answered that he could not because the palace was so far away. 
Sidi c Ali told him to close his eyes. When he opened them, he found 
himself in front of Moulay IsmaiYs palace. *Abdelhaqq called him in 
and offered him some buttermilk. Sidi Ahmed drank a whole pail of 



THE LEGENDS 


33 


buttermilk , then a second , and then a third. Nothing happened. His 
stomach did not explode. It was not even swollen. A fourth pail still 
remained. Sidi Ahmed offered it to 'Abdelhaqq, who drank it. Then 
Sidi Ahmed shot him with an arrow. 

The soldiers ran to Moulay Ismail and told him what Sidi Ahmed had 
done. Moulay Ismail ordered the soldiers to get both Sidi 'Ali and Sidi 
Ahmed. Both men refused to come. When Moulay Ismail heard this , 
he ordered his soldiers to bring them by force , to drag them all the 
way if necessary. The soldiers returned to Sidi 'Ali , who asked them 
if the sultan wanted him to come alone or with his friends. The sol¬ 
diers returned to Moulay Ismail to ask whether or not he wanted Sidi 
'Ali alone or with his friends. Moulay Ismail answered , “Tell him to 
come with his monkeys and dogs.” The soldiers returned and told Sidi 
'Ali to bring his “monkeys and dogs'' As Sidi 'Ali started to get up } the 
mountain bent over and rocks began to slide down. The soldiers told 
Sidi 'Ali not to move. They would have to ask the king what was to 
be done. Moulay Ismail answered, “If he has so much baraka , let him 
take my oil to Mecca” The road to Mecca was blocked at the time , 
and Moulay Ismail could not send oil there. 

The soldiers returned to Sidi 'Ali and told him what the sultan had 
said. The saint instructed them to bring him the oil , which they brought 
in skin bags and poured down the throat of the reclining saint. Then , 
as they were on their way back to the palace to fetch more oil, a mes¬ 
senger arrived from Mecca to tell Moulay Ismail that the oil had 
already arrived safely. 

Lalla 'Auda, one of Moulay IsmaiPs wives, heard about Sidi 'Alps 
miracles and wanted to visit him because he had so much baraka. She 
had not had any children and hoped that the saint would be able to 
help her. Motday Ismail sent her with so?ne slaves and told them that 
he would have both her head and theirs if they were not all back by 
sunset. When Lalla 'Auda arrived , she found Sidi 'Ali asleep. She woke 
him and explained that she was in a hurry. Sidi 'Ali told her not to 
worry. She had plenty of time, he said. Lalla c Auda remained at the 
saint's until there were only fifteen minutes to sunset. Then Sidi 'Ali 
pulled Moulay IsmaiPs hand all the way from Meknes to massage her 
belly so that she would have children. When one of the slaves saw Sidi 
'•Ali massaging his master's wife's belly, he started to protest , but Sidi 
'Ali told him to shut up or he would have him shot. Then Sidi 'Ali 
dismissed Lalla 'Auda, telling her to say when she arrived in Meknes , 
“O, Sun, set and may Allah be thanked.” 

When she arrived in Mek?ies } she found the whole city in an uproar. 




34 


THE HAM ADS HA 


Everyone was wondering why the sun had not set. Moiday Ismail him¬ 
self was staring at the sun! Lalla c Auda asked him what was wrong and 
then told him what Sidi c Ali had said. He ordered her to repeat the 
words, and then the sun set. Moulay Ismail then turned to his slave and 
asked him what he had done when Sidi *Ali had massaged Lalla 'Audds 
belly. The slave answered that he had tried to stop him. Moulay Ismail 
cried out, “You fool. It was my hand” He ordered his soldiers to load 
their guns and shoot the slave. This they did. 

Sidi c Ali turned to his side and found Sidi Ahmed there. He told him 
to go to the Sudan to get the hal. Sidi Ahmed answered that it was far 
away. It was at least a six-months* journey. Sidi c Ali told him to close 
his eyes, and when he opened them he found himself in front of the 
palace of the ki?ig of the Sudan. Sidi Ahmed entered the palace, where, 
praise be to God, the soldiers were all asleep. There he found an 
c awwad [a short reed flute], a daff [a square flat drum], and 'A'isha 
Qandisha [a she-demon, or jinniyya, to whom the Hamadsha are de¬ 
voted]. Sidi Ahmed took them with him. Then the soldiers and the 
king awoke. The king asked them where his 'awwad, his daff, and 
'A'isha Qandisha had gone. The soldiers said they did not know a?id 
assured him that no one had entered while he was asleep. The king 
told them that it must have been Sidi Alymed who did it, and ordered 
them to follow the saint to the c Ayn Kabir. 

When Sidi Ahmed found out that the king knew it was he, he sent 
a message by pigeon to Sidi 'Ali to pray for him so that he would not 
be taken prisoner. Because of the baraka of Sidi 'Ali, the soldiers were 
all turned into frogs. When the king died, his body was taken to the 
*Ayn er-Rjal at Moulay Idriss and buried. After his burial, whenever 
the inhabitants of Moulay Idriss prepared couscous or tajin [stew], 
frogs jumped from their plates. They brought a sacrifice to Moulay 
Idriss, who woke and told them that they had brought the sacrifice to 
the wrong person, that they should go to Sidi c Ali ben Hamdush. The 
townspeople then went to Sidi 'Ali and told him that frogs were jump¬ 
ing all over the place. Sidi c Ali asked them who had sent them, and 
when he learned that it was Moulay Idriss, he instructed them to return 
to their village. There they would find a man dressed in a darbala [a 
tattered and patched cloak], sleepmg. They were to ask him his ad¬ 
vice, and he would tell them what to do. They found the man, woke 
him, and asked him if it was he who had let out all the frogs. The man 
asked them who had sent them, and when they told him that it was 
Sidi 'Ali, he told them that they should work the hal of Sidi 'Ali. The 
people agreed to do this and have made sacrifices to Sidi 'Ali ever since. 



THE LEGENDS 


35 


One day Sidi Ahmed went to fetch water for Sidi c Ali. When he 
returned , he found the saint dead. He began to hit his head with his 
hands , crying “ Allah/ The saint is dead!” Since that time there has been 
the gwal [drum] and the head-slashing. 

Explication of the Legends 

I will now consider the various sections of the legend in some detail. 

Sidi c Ali was caid to Moulay Ismail. Moulay Ismail was a ferocious man 
who cut off the heads of liars. One day he ordered Sidi c Ali to fetch 
Bu'abid Sharqi. Sidi c Ali went to Sharqi with his troops. Sharqi asked Sidi 
e Ali if he had come to get him. Sidi c Ali answered that he had. Sharqi then 
gave Sidi c Ali a hand-mill and told him to turn it. Sidi c Ali turned the mill, 
and when he stopped, the mill continued to turn on its own. Sidi c Ali threw 
some grain into the mill. A black woman came up to him and exclaimed that 
the mill was grinding the grain by itself. Sidi c Ali told the woman to shut 
up and cursed her blind. When Sharqi saw that the woman was in fact 
blind, he threw Sidi c Ali out and immediately fell ill and vomited into a pail. 
Then he ordered the blind woman to get rid of the vomit. Feeling her way 
along the wall, she took the pail to the door, whereupon Sidi c Ali, who had 
been hiding there, grabbed the pail and drank all of the vomit. This is how 
he got the remainder of his baraka from Bu'abid Sharqi. 

The legend begins typically in the middle of Sidi ‘Ali’s life, at the 
time when he is about to obtain the “remainder” of his baraka. We are 
not told where Sidi e Ali came from or how he became Moulay Ismail’s 
caid. When questioned about this, the muqaddim explained that Sidi 
c Ali was a descendant of Moulay Idriss (d. 791 a.d.) who is popularly 
considered to have brought Islam to Morocco; he founded the first 
Moroccan dynasty. (Moulay Idriss is also buried on the Jebel Zcrhoun, 
midway between Fez and Meknes, and just above the Roman ruins of 
Volubilis.) The muqaddim then corrected himself and explained that 
Sidi c Ali was, in fact, bom in Mecca and came West with Moulay 
Idriss and Moulay Idriss’ freedman Moulay Rachid, both of whom 
stopped at the village of Moulay Idriss while Sidi e Ali went on to Beni 
Rachid. He could not explain how Sidi c Ali became Moulay Ismail’s 
caid. 

Although the lives of Sidi c Ali and the sultan Moulay Ismail (d. 
1727), the founder of the city of Meknes, did overlap, it should be 
noted that Moulay Ismail is a stock character in many of the hagio- 
graphic legends. He is reputed for his extreme and arbitrary cruelty, 




36 


THE HAMADSHA 


which is usually expressed in terms of promiscuous beheadings. Often 
—and later in the muqaddim’s account itself—he is seen as a protagonist 
to the saint—an expression, perhaps, of an incipient schism between the 
secular and the sacred order of things, or simply Aloulay Ismail’s em¬ 
phasis on the strong-man aspect of his role as sultan, rather than on the 
holy-man aspect (cf. Geertz 1968:53). It may also be proof of Sidi 
‘Ali’s power over as virile a figure as the sultan (who beheads his an¬ 
tagonists). That Sidi £ Ali is seen as an emissary of Aloulay Ismail is 
unusual. The muqaddim may have wished to emphasize the saint’s im¬ 
portance and therefore identified him with Aloulay Ismail, whose 
strength he admired. Perhaps he wanted to begin his tale as the heroic 
tales of kings and warriors recounted by professional storytellers in 
marketplaces and cafes begin. 

Interpretation of the episode is difficult. Unlike other versions of 
the passage of baraka from Bu'abid Sharqi to Sidi £ Ali, which conceive 
of it as a friendly gift, the muqaddim has emphasized the competitive 
element. But in his interpretation he suggests—and thereby contradicts 
himself—a more friendly aspect. Originally he described the episode 
as a competition between the two saints. First Bu £ abid Sharqi performs 
a miracle ( burhan ): he makes the mill turn on its own. (The muqad¬ 
dim later added that it was only after Sharqi had touched Sidi ‘All’s 
arm that the mill turned on its own.) Then Sidi £ Ali also performs a 
burhan: he blinds the black woman. The muqaddim saw no wrong in 
this. When I asked him why Sidi £ Ali blinded the woman, he answered: 
“Well, for example, there is an old man sitting down. A little boy 
comes up to him and asks what he is doing. The old man tells the boy 
to scram. Then the old man curses the child.” Sidi c Ali was presumably 
irritated at the black woman for bothering him, for pointing out the 
strength of Sharqi’s baraka, and he cursed her. Still, even when pressed, 
the muqaddim could not conceive of the two saints being jealous of 
each other. 

Sidi c Ali did not stay at Sharqi’s in order to deceive the saint. He 
remained, the muqaddim explained by adding a new scene to his ver¬ 
sion, because his warriors had left with the horses. They thought their 
caid was dead. Sharqi himself knew that Sidi £ AJi was at the door wait¬ 
ing, and would drink the vomit. He had seen that Sidi c Ali had a lot 
of baraka and wanted to give him his own baraka before he died. 

The conception of the passage of baraka—the master-disciple rela¬ 
tionship—by the oral incorporation of vomit, polluted water, bath¬ 
water and other foul materials occurs frequently in the legends of vari¬ 
ous Aloroccan saints. In another version of the same story, Sidi £ Ali 



THE LEGENDS 


37 


comes to work for Bu'abid Sharqi as a water-carrier and not only drinks 
the water that drips from Sharqi’s water-sack—Sharqi himself is a 
water-carrier who takes no money for the water he carries—but drinks 
the water with which the saint has washed himself before his prayers. 
Sometimes, however, the baraka is passed to the saint in the form of 
bread—bread being a recurrent symbol for baraka in both legend and 
ritual. Sidi c Ali works in disguise for Sharqi because “Sharqi had de¬ 
stroyed Sidi 'Ali’s papers” as a water-carrier; and he makes the mills 
of eighty of Sharqi’s women turn on their own. A Negro woman tells 
Sharqi; Sidi c Ali blinds the woman; and Sharqi, recognizing Sidi e Ali 
as a saint, gives him two loaves of bread and tells him that he will obtain 
the rest at Moulay Idriss. At Moulay Idriss, Sidi e Ali encounters forty 
saints guarding the bread for him. 

Then Sidi e Ali began a long trip and arrived at the e Ayn Kabir and sat down. 

The muqaddim was ignorant of Sidi ‘Ali’s adventures as the saint 
traveled from Sharqi’s to the ‘Ayn Kabir. He simply stated that the 
saint wandered from village to village, gathering followers and found¬ 
ing lodges. The ‘Ayn Kabir, or Big Spring, is a spring that flows from 
the cliffs just below Sidi ‘Ali’s tomb in Beni Rachid and is believed to 
contain much baraka. It is visited by pilgrims to Sidi ‘Ali’s tomb who 
either drink the water, take a little home, or bathe in its waters. The 
muqaddim explained that the water was not filled with baraka until 
the saint had bathed in it. 

Although the muqaddim himself knew nothing of what happened 
to Sidi ‘Ali on his trip to the ‘Ayn Kabir, a number of episodes con¬ 
cerning this period were told to me by other informants. They are all 
concerned with Sidi ‘Ali’s acquisition of baraka or with his recognition 
as a saint. 

According to one version, Sidi ‘Ali drank all of the water in Sharqi’s 
water-bag. The saint had been in the habit of giving a little of the 
water—a little of his baraka—to forty people each year, but Sidi ‘Ali 
took it all for himself. Sharqi ordered his men to chase Sidi ‘Ali all the 
way to Tadla, if need be, and assured them that they would die if Sidi 
‘Ali ever returned to Marrakech, where Sharqi was living at the time. 
Sidi ‘Ali fled to Fez to stay with Sidi 1-Husayn, who was dying and 
surrounded by forty saints. Sidi ‘Ali stood at the door and drank Sidi 
1-Husayn’s vomit. When the saint heard about this from his serving 
women, he cried praise to Allah and died in peace as the other saints 
filed out of his room. 



38 


THE HAM ADS HA 


Sidi e Ali then went to the Qarwiyyin University at Fez—or so this 
particular version continues—and sat down in a comer praying and 
spitting. He did not move for many years. The scholars of the univer¬ 
sity grew suspicious. One of them, a member of the well-known ben 
Suda family of Fez, decided to chase him out; but somehow, every time 
he approached the saint, he would forget his purpose. One night ben 
Suda dreamed that the sultan, the royal guard, and the army were in 
the university courtyard. The sultan was holding out a loaf of bread 
to the scholars and his warriors; but each time one of them tried to take 
the bread, he was curiously stung. Finally, when all had tried, only Sidi 
'Ali remained. He stood up, took the bread, and left the university for 
Beni Rachid. When ben Suda awoke, he rushed to the university and 
discovered Sidi 'Ali’s corner empty. Then he knew that Sidi 'Ali was 
indeed a saint. 1 

Sidi e Ali also spent time with Sidi Qasim, it is said, in the town of 
Sidi Kacem in the Gharb, where he worked as a gardener for four or 
five years and never touched any of his master’s fruits. He fasted con¬ 
tinually. One day Sidi Qasim was entertaining guests and asked the 
gardener to bring him some good pomegranates. Sidi e Ali, who never 
touched the fruit, could not tell a good pomegranate from a bad one 
and simply brought in the biggest pomegranate he could find. The 
fruit was spoiled. Sidi Qasim then called his gardener and asked him 
why he had chosen a bad pomegranate. When the saint explained that 
he could not distinguish because he had never tasted one, Sidi Qasim 
realized that Sidi e Ali was in fact a saint. 

Someone came up to Sidi c Ali. His name was Sidi Ahmed Dghughi. Sidi 
e Ali told him to heat some water so that he could wash for his prayers. 
Sidi Ahmed asked if God would give him baraka if he did this. Sidi c Ali 
assured him that He would. 

The muqaddim expanded this episode in his explanation. When Sidi 
'Ali arived at the e Ayn Kabir, he was visited by people from all over 
Morocco. Once a particularly large group of pilgrims arrived. Sidi c Ali 
saw Sidi Ahmed in their midst and called him out to work for him. 
He saw that he was devout; his heart told him that Sidi Ahmed should 
work for him. 

The descendents of Sidi Ahmed Dghughi are always careful to point 
out that Sidi Ahmed was not Sidi 'Ali’s servant or slave but a friend 

1 My informants were not so much awed by the fact that Sidi e Ali was able 
to take the bread as by his ability to remain immobile for years. 



THE LEGENDS 


39 


and an equal. Sidi c Ali is often seen as dependent on Sidi Ahmed in the 
legends told by them. Once, for example, when members of Sidi 
Qasim’s family were jealous of Sidi c Ali, they tried to poison him with 
a couscous made with snakes . 2 (Snake meat is considered poisonous in 
Morocco.) Sidi Ahmed warned his master. As it turned out, Sidi c Ali 
ate the couscous anyway, and the skins of his enemies began to peel off. 
Most of the legends about Sidi Ahmed are concerned with the missions 
he performed for Sidi ‘Ali, although his descendants have recounted 
several which do not involve Sidi c Ali. Two of these are about Sidi 
Ahmed’s conflicts with the canonists of Fez (the Sufi versus the c ulama). 

While Sidi Ahmed was in Fez, the scholars asked him to give a dis¬ 
course on prayer and other religious matters. Sidi Ahmed said, “All that 
is necessary is a silo of wheat and a silo of barley” The scholars did 
not understand. Sidi Alrmed then invited them into his house, which 
he locked and in which there was no food. For three or four days the 
men talked of prayer and other religious matters. Frnally they began 
to ask for food, but Sidi Ahmed simply replied: “Continue your dis¬ 
cussion” Finally they became so hungry that they could not even say 
their prayers. Then they realized the wisdom of Sidi Ahmed's words. 
The proverb, “Without bread there is neither prayer nor devotion” 
comes from this. 

The scholars of Fez did not believe that Sidi Ahmed was a saint. “You 
say” they said, “that you can make a camel of a man. Show us how this 
is possible.” Sidi Ahmed asked them to bring prickly pears which were 
still covered with spines. One of his followers who was in ecstasy 
(hal) devoured the pears, spines, skin, and all, and cried out that he 
was still hungry and wanted bread. The scholars said that the bread 
was still in the oven. Sidi Ahmed asked them to lead him and the 
“corner to the oven. The “camel” ate all of the bread as the scholars 
waited for the two men to come out of the oven-house, and then they 
escaped by a back door of the oven, which opened miraculously. When 
the scholars discovered this, they pursued the two me?t all the way to 
Mehdiya near Kenitra to present their apologies. 

This episode refers to certain followers of Sidi Ahmed (and Sidi c Ali) 
who act like camels and eat the spines or skins of prickly pears when in 
trance. The reader should also be reminded that bread is a symbol of 
baraka. 

*In other versions the poisoners are Sidi ‘Ali’s brothers. 



40 


THE HAMADSHA 


The third episode concerns Sidi Ahmed and Moulay Ismail, and 
again reflects the conflict between saint and sultan that was noted 
earlier. 

A lot of people considered Sidi Ahmed a magician because he would 
become like a camel and eat spiny cactus when he was in hal. The 
people told this to Moulay Ismail, who went to see him and challenged 
him to a battle. Many of Sidi Ahmed's followers were afraid of Moulay 
Ismail and did not follow him. The two “ armies ” met at a bog between 
Fez and Meknes, where the two leaders decided to fight it out alone . 
As they touched hands before the duel, the earth suddenly opened up, 
forming a huge chasm around Moulay Ismail, who then realized that 
Sidi Ahmed was a saint and surrendered to him. 

I have heard this story told of Sidi 1-Hasan 1-Yussi of Sefrou too. 

At this time the son of Moulay Ismail, ‘Abdelhaqq, was in the habit of 
shooting arrows into passersby. Before killing them, he would invite them 
in to drink some buttermilk. Then he would shoot them. Sidi 4 Ali told Sidi 
Ahmed to go to the palace to do something about this. Sidi Ahmed answered 
that he could not because the palace was so far away. Sidi *Ali told him to 
close his eyes. When he opened them, he found himself in front of Moulay 
Ismail’s palace. 'Abdelhaqq called him in and offered him some buttermilk. 
Sidi Ahmed drank a whole pail of buttermilk, then a second, and then a 
third. Nothing happened. His stomach did not explode. It was not even 
swollen. A fourth pail still remained. Sidi Ahmed offered it to ‘Abdelhaqq 
who drank it. Then Sidi Ahmed shot him with an arrow. 

The soldiers ran to Moulay Ismail and told him what Sidi Ahmed had 
done. Moulay Ismail ordered the soldiers to get both Sidi e Ali and Sidi 
Ahmed. Both men refused to come. When Moulay Ismail heard this, he 
ordered his soldiers to bring them by force, to drag them all the way if 
necessary. The soldiers returned to Sidi c Ali, who asked them if the sultan 
wanted him to come alone or with his friends. The soldiers returned to 
Moulay Ismail to ask whether or not he wanted Sidi ‘Ali alone or with his 
friends. Moulay Ismail answered, “Tell him to come with his monkeys and 
dogs.” The soldiers returned and told Sidi c Ali to bring his “monkeys and 
dogs.” As Sidi e Ali started to get up, the mountain bent over and rocks 
began to slide down. The soldiers told Sidi c Ali not to move. They would 
have to ask the king what was to be done. Moulay Ismail answered, “If he 
has so much baraka, let him take my oil to Mecca.” The road to Mecca 
was blocked at the time, and Moulay Ismail could not send oil there. The 
soldiers returned to Sidi c Ali and told him what the sultan had said. The 
saint instructed them to bring the oil, which they brought in skin bags and 



THE LEGENDS 


41 


poured down die throat of the reclining saint. Then, as they were on their 
way back to the palace to fetch more oil, a messenger arrived from Mecca 
to tell Moulay Ismail that the oil had already arrived safely. 

The muqaddim explained that since ‘Abdelhaqq was the son of 
Moulay Ismail, there was no one who could control him. He found 
this quite understandable. ‘Abdelhaqq gave his victims buttermilk be¬ 
cause it pleased him to see their stomachs swell and then to see the milk 
come squirting out of their pierced stomachs. (Buttermilk is said to 
contain baraka, and offering it is considered a sign of hospitality. It is 
popularly believed to make the stomach swell and even explode if one 
drinks too much of it.) ‘Abdelhaqq did not shoot Sidi Ahmed because, 
thanks to Sidi c Ali, his stomach did not swell. The sultan’s son had to 
accept Sidi Ahmed’s challenge because he was surrounded by viziers 
and courtiers. It would have been shameful (hshuma) to refuse. Be¬ 
sides, ‘Abdelhaqq thought that he himself had so much baraka that his 
stomach would not swell either. Sidi Ahmed killed ‘Abdelhaqq because 
—the muqaddim quoted the proverb—“People who kill others for no 
reason will be killed by an emissary of God.” The courtiers did not 
take Sidi Ahmed prisoner because he had so much baraka. 

The sultan called for both Sidi ‘Ali and Sidi Ahmed, the muqaddim 
continued, because they were responsible for the death of his son. He 
was angry and insulted Sidi ‘Ali’s friends by calling them “monkeys 
and dogs.” He wanted to see what miracles (burham) Sidi ‘Ali could 
perform, and gave him the impossible task of getting oil to Mecca. 
When Sidi ‘Ali drank the oil, it was not he but “others” who took the 
oil to Mecca. The muqaddim seemed uncertain as to whether the 
“others” were angels (malaika) or jnun. Finally he decided on angels. 8 
The messenger who informed Moulay Ismail of the arrival of the oil 
was also an “invisible.” 

This tale of Sidi ‘Ali’s drinking Moulay Ismail’s oil in order to “send” 
it to Mecca is popular in the area of Meknes. In another version, 
Moulay Ismail calls for Sidi Ahmed; Sidi ‘Ali takes his place because 
he is not married and Sidi Ahmed is. When Sidi ‘Ali arrives at the 
palace, the sultan tells him that he must produce fifteen jugs of oil to 
be sent to Mecca by the end of the week or he will lose his head. Sidi 
‘Ali returns at the end of the week and drinks the sultan’s and his 
family’s bath water and vomits it out into the waiting jugs as oil. 

* This confusion is not as great as the English glosses for “malaika” and “jnun” 
suggest. The jnun are not necessarily evil and harmful. Many of my informants 
confused angels and jnun in their tales and associations (cf. Chelhod 1964 : 67 - 92 ). 



42 


THE HAMADSHA 


Lalla e Auda, one of Moulay Ismail’s wives, heard about Sidi e Ali’s miracles 
and wanted to visit him because he had so much baraka. She had not had 
any children and hoped that the saint would be able to help her. Moulay 
Ismail sent her with some slaves and told them that he would have both 
her head and theirs if they were not all back by sunset. When Lalla c Auda 
arrived, she found Sidi c Ali asleep. She woke him and explained that she 
was in a hurry. Sidi e Ali told her not to worry. She had plenty of time, he 
said. Lalla e Auda remained at the saint’s until there were only fifteen min¬ 
utes to sunset. Then Sidi c Ali pulled Moulay Ismail’s hand all the way from 
Meknes to massage her belly so that she would have children. When one of 
the slaves saw Sidi c AIi massaging his master’s wife’s belly, he started to 
protest, but Sidi e Ali told him to shut up or he would have him shot. Then 
Sidi e Ali dismissed Lalla c Auda, telling her to say when she arrived in 
Meknes, “O, Sun, set and may Allah be thanked.” 4 

When she arrived in Meknes, she found the whole city in an uproar. 
Everyone was wondering why the sun had not set. Moulay Ismail himself 
was staring at the sun! Lalla c Auda asked him what was wrong and then 
told him what Sidi e Ali had said. He ordered her to repeat the words, and 
then the sun set. Moulay Ismail then turned to his slave and asked him 
what he had done when Sidi e Ali had massaged Lalla 'Auda’s belly. The 
slave answered that he had tried to stop him. Moulay Ismail cried out, “You 
fool. It was my hand.” He ordered his soldiers to load their guns and shoot 
the slave. This they did. 

This episode is one of the most popular in the life of Sidi c Ali and 
accounts for his title “Guide of the Sun” (gawwad sh-shamsh) or “Caid 
of the Sun” (qaid sh-shermsh). I have been told that if one looked at 
Sidi ‘Ali’s knees, one could see the sun shining both by day and by 
night. 

The muqaddim corrected himself in his explanation. Lalla e Auda 
went to Sidi c Ali with only one slave whose name was Brahim. Lalla 
e Auda wanted a son so that she could remain Moulay Ismail’s favorite 
wife. ( c Auda means mare in Moroccan Arabic.) Sidi e Ali could not 
touch the woman’s belly with his own hand, because then Moulay 
Ismail would never be able to touch her himself. She would have be¬ 
come taboo ( haram ). The descendants of the saint still massage the 
bellies of barren women, the muqaddim explained, and they receive 
gifts every day from women who have then had children. Moulay Is¬ 
mail had Brahim killed, even though the slave did not know that Sidi 
c Ali was in fact massaging Lalla ‘Auda’s belly with his master’s hand, 
because he had questioned the saint. The muqaddim seemed surprised 

‘According to the version of this legend cited by Herber, Lalla c Auda was to 
say: “Continue your journey, O Sun, by the power of Allah” (1923:224). 



THE LEGENDS 


43 


when I asked him if the fact that the slave did not know it was Moulay 
Ismail’s hand made any difference. He assured me it did not. 

Other versions of the episode suggest that Moulay Ismail was very 
discontented with Lalla c Auda because she had not borne him a son 
and threatened to have her head if she did not conceive the following 
day. That night, she dreamed of Sidi c Ali, and she visited him the fol¬ 
lowing day. Sidi c Ali did not massage her belly but gave her a scarf 
which she was to wear against her belly, under her clothes. 

Sidi e Ali turned to his side and found Sidi Ahmed there. He told him to go 
to the Sudan to get the hal. Sidi Ahmed answered that it was far away. It 
was at least a six-month’s journey. Sidi c Ali told him to close his eyes, and 
when he opened them he found himself in front of the palace of the king 
of the Sudan. Sidi Ahmed entered the palace, where, praise be to God, the 
soldiers were all asleep. There he found an 'awwad, a daff, and c Aisha 
Qandisha. Sidi Ahmed took them with him. Then the soldiers and the king 
awoke. The king asked them where his c awwad, his daff, and e AIsha 
Qandisha had gone. The soldiers said they did not know and assured him 
that no one had entered while he was asleep. The king told them that it 
must have been Sidi Ahmed who did it and ordered diem to follow the 
saint to the e Ayn Kabir. 

When Sidi Ahmed found out that the king knew it was he, he sent a 
message by pigeon to Sidi e Ali to pray for him so that he would not be 
taken prisoner. Because of the baraka of Sidi c Ali, the soldiers were all 
turned into frogs. When the king died, his body was taken to the e Ayn 
er-Rjal at Moulay Idriss and buried. After his burial, whenever the in¬ 
habitants of Moulay Idriss prepared couscous or tajin [stew], frogs jumped 
from their plates. They brought a sacrifice to Moulay Idriss, who woke and 
told them that they had brought the sacrifice to the wrong person, that 
they should go to Sidi e Ali ben Hamdush. The townspeople then went to 
Sidi e Ali and told him that frogs were jumping all over the place. Sidi e Ali 
asked them who had sent them, and when he learned that it was Moulay 
Idriss, he instructed them to return to their village. There they would find 
a man dressed in a darbaJa, sleeping. They were to ask him his advice, and 
he would tell them what to do. They found the man, woke him, and asked 

him If it was he who had let out all the frogs. The man asked them who had 

sent them, and when they told him it was Sidi c Ali, he told them that they 
should work the hal of Sidi e Ali. The people agreed to do this and have 
made sacrifices to Sidi e Ali ever since. 

Sidi Ahmed’s trip to the Sudan is considered to be the most important 
episode in the legendary cycle, for it explains both the origin of the 
hal, or ecstatic trance, into which the Hamadsha work themselves, and 
the arrival of 'Ai'sha Qandisha. 




44 


THE HAMADSHA 


‘Aisha Qandisha is a she-demon, sometimes called a jinmyya , or fe¬ 
male jinn, sometimes an c afrita, or giant jinn-like creature, and some¬ 
times she is considered to be quite different from either of these. She is 
said to appear sometimes as a beautiful woman and sometimes as an old 
hag, but always with the feet of a camel or some other hoofed animal. 
Despite a widespread belief in ‘A'isha Qandisha among the northern 
Moroccan Arabs, the Hamadsha, as we shall see, are her special dev¬ 
otees. They hold her responsible for their trance. Although her Su¬ 
danese origin is almost universally accepted, it has been claimed that 
she was in fact a woman and Sidi ‘Ali’s slave and that when she was 
buried she suddenly disappeared and was heard laughing. “I am here,” 
she called out. “I am dead. I’ve been carried away by my mcmiluk 
(jinn).” Some hold her to be the daughter of Sidi Shamharush, the king 
of the jnun; and on at least one occasion I was told that her mother 
was human and her father Ighud, the shepherd of the wind ( rih ), who 
carried her mother off to the forest. This is said to explain why she has 
both a human name, ‘A'isha, and a devil’s name, Qandisha 5 (Haimer 
1968 : 52 ). 

The muqaddim’s version of the episode is the most complete I have 
heard; the frog section is unique to him. (The jnun are often said to 
take the form of frogs.) Sometimes Sidi Ahmed is said to have gone to 

the Sudan by way of Mecca. The Sudan refers to the area immediately 
south of the Sahara and is considered by many Moroccans to be the 
home of the hal. I have been told that Sidi Ahmed found ‘Ai'sha Qan¬ 
disha in a place called Sakqia al-Hamra, probably in the northern part 
of the Spanish Sahara. In another version, it is the jnun, alerted by the 
whinny of Sidi Ahmed’s horse, who pursue him. He blinds them by 
throwing tar ( qatran) in their eyes, and then sand. This explains why 
pilgrims to Sidi Ahmed’s tomb take back a little earth {tuba) with 
them as protection from the jnun. The instruments referred to by the 
muqaddim are not used by the Hamadsha today, but the muqaddim 
claimed they were once played. 6 Often Sidi Ahmed is said to have 
obtained ‘Ai'sha not from the king of the Sudan but by pulling open a 
cord which held a bundle of wood together 7 and to have returned to 
Sidi ‘Ali followed by thousands of devotees including the most famous 
saints of the land. In most versions Sidi ‘Ali has died. He then makes 

0 See Chapter Eight for a more detailed account of c ATsha Qandisha. See Jones 
(1951:279-286) for similar myths. 

•See Garcia-Barriuso (1941:105) for a description of the daff ( deff ). 

T The Hamadsha occasionally beat themselves with bundles of sticks (an example 
of which may be seen in the Musee de l’Homme in Paris), along with their other 
instruments of self-mutilation. 



THE LEGENDS 


45 


a pact with 'Aisha: she must never come near him. The man in the 
darbala is claimed to have been Sidi 'Ali in disguise. He could not very 
well himself have told the townspeople of Moulay Idriss to devote 
themselves to his hal. 

One day Sidi Ahmed went to fetch water for Sidi e Ali. When he returned, 
he found the saint dead. He began to hit his head with his hands, crying 
“Allah! The saint is dead!” Since that time there has been the gwal and the 
head-slashing. 

This version of the legend explains at least why the followers of Sidi 
Ahmed slash their heads—self-mutilation during mourning is not un¬ 
common in Morocco. It has been claimed that only Sidi Ahmed’s fol¬ 
lowers abandon themselves to such practices (Aubin 1904:431), but in 
fact the followers of both saints do, and other legends suggest that Sidi 
‘Ali himself also did so. Cat (1898:401) reports that the Hamadsha 
slash their heads in commemoration of Sidi ‘Ali’s putting his head back 
together after it had been split open by a shell. Sidi ‘Ali is also said to 
have pounded his chest while he said his beads (tasbih). The beads 
were attached to a tree and every time a bead was passed the saint 
would beat his chest (Herber 1923:225). I have heard neither of these 
legends. 

The general belief is that Sidi Ahmed, finding his master dead, first 
washed and buried him and then climbed to the top of the Jebel Zer- 
houn, where he became so upset that he slashed his head with an ax 
as he called out the names of all the saints. The people tried to calm him. 
“You are lucky to be alive,” they said. “You are lucky to have children. 
These children will carry on your blood and your baraka. You must 
not be angry or sad. You will always have children. They will always 
replace you.” 

They told him that before Sidi c Ali died, he said “For you Thursday, 
for me Friday,” by which he meant that Sidi Ahmed had inherited his 
baraka and would be visited first, on Thursdays, at the annual pilgrim¬ 
age, or vnisem. It has been said that the first musem took place when 
Moulay Ismail came to Sidi 'Ali’s tomb on the last day of the mourning 
period (forty days after the saint’s death). Sidi Ahmed lived for several 
years—he would slash his head whenever he was falling asleep as he 
prayed—and when he died, some say among the Beni Hsen, he was put 
on a mule, as he requested, and carried as far as the mule would go. 
The mule collapsed at Beni Ouarad, a mile from his master’s tomb, and 
it was here that he was buried and that some of his descendants settled. 



46 


THE HAMADSHA 


Although I was personally unable to learn of the origin of the instru¬ 
ments with which the Hamadsha slash their heads, Herber gives the 
following two legends, which account for the use of the single- and 
double-bladed axes the Hamadsha traditionally employed and still use 
during their annual pilgrimage. The first of these concerns Sidi Rahal, 
a saint who is buried southeast of Marrakech in the town named after 
him and who is the founder of the Rahaliyyin brotherhood, whose 
members are reputed for drinking boiling water when in an entranced 
state, and another saint, Sidi Ahmed 1-Arusi. 

Sidi Rahal was living on a mountain near Marrakech; he was a her¬ 
maphrodite and had with him two chaste women who served him. The 
sultan , having heard his saintliness praised, wanted to see him. He sent 
two of his people to fetch him: one was Sidi *Ali ben Hamdush y the 
other Sidi Ahmed l-Arusi. But these two were moved by his piety to 
the point that they attached themselves to him and did not want to 
return to Marrakech. The sultan sent his soldiers for them y and they 
were condemned to death. As they arrived at the place of execution , 
Sidi Rahal appeared in the skies y holding an axe which filled the spec¬ 
tators with dread. He gave it to Sidi c Ali ben Hamdush and told him 
thatf thanks to it y he should have no fear of anyone at all. Then he took 
Ahmed l-Arusi by the belt and carried him off in flight. The belt tore y 
l-Arusi fell into the [Valley of the] Sous y where he ended his life in 
retirement. Sidi Rahal returned to his hermitage; Sidi ‘Ali retired to 
the Xerhoun y where he bequeathed his axe to his followers and died. 
(Herber 1923:225 , fn. 1) 

The origin of this legend is not given. The following one was told to 
the author by an interpreter in Rabat. 

Sidi ‘Ali ben Hamdush retired to the mountain y several kilometers 
from Meknes, in order to lead a solitary life and to pray to God. Sev¬ 
eral of his companions followed him. They lived on alms and nourished 
themselves principally on the roasted heads of sheep , the bones of which 
soon formed a big pile in front of the cave in which Sidi c Ali and his 
companions lived. Every day y at sunrise and after prayers y Sidi ‘Alps 
disciples gathered around him and began their dances which they did 
not stop except during the hours of prayer and at meals. 

One day , not seeing the arrival of their leader , the brothers were 
taken with fear; none dared to go to learn the news of which they had 
a foreboding; they took counsel and chose Sidi ‘Alps favorite disciple , 



THE LEGENDS 


47 


who sat to his right during the recitation to God. All trembling and 
soaking with sweat, the elected one went to the cave of their venerated 
leader; he reappeared soon, his mouth wide open, his eyes out of their 
sockets, and threw himself on the sheep bones. He took a jawbone and 
hit his head saying: “Allah! Allah! There is no God but AllahP ’ His 
companions, understanding the truth, hit themselves in imitation of 
their new leader. Hence, the origin of the ax which “is almost in the 
shape of a jawbone.” (Herber 1923:225, fn. 1) 


Interpretation of the Legends 

All of these legends may be analyzed on many different levels and 
from many different theoretical perspectives. Some of the legends, for 
example, appear to be explanations for certain ritual practices, such as 
taking home a little earth from the tomb of Sidi Ahmed or slashing 
one’s head with an axe “shaped like a jawbone.” Others appear to reflect 
primarily economic tensions, as between the descendants and followers 
of Sidi c Ali and Sidi Ahmed; politico-religious tensions, as between 
Moulay Ismail and the saints; differences in theological perspective, as 
between the canonists of Fez and the saints of the Zerhoun; or notions 
of the relative importance, the hierarchization, of saints—Sidi Ahmed, 
Sidi c Ali, and Moulay Idriss. I will be concerned, however, only with 
those elements of the legends which are germane to an understanding 
of the Hamadsha therapeutic system. 8 On the one hand, I will focus 
on the image of the saint as expressed in his existential stance vis-a-vis 
certain significant elements within his life; and on the other hand, on 
the passage of baraka —the curative force—from saint to saint. 

Despite the fact that there is no fixed order of events in the legends 
in general, a thematic analysis of Muqaddim Touijer’s version of the 
life of Sidi c Ali reveals a definite development, at the most literal level 
of analysis, from the earliest episode, which is concerned with Sidi ‘Ali’s 
acquisition of baraka, to the final one, which deals with the passage of 
liis baraka to Sidi Ahmed. Sidi Ahmed is recognized as a saint and is 
responsible for the head-slashing of the Hamadsha. The intervening 
episodes are concerned with Sidi ‘Ali’s recognition as a saint by various 
other saints, canonists, and political figures such as Moulay Ismail, and 
with his acquisition of disciples, such as Sidi Ahmed, and devotees, such 
as the people of Moulay Idriss. 

8 At this stage of investigation, my choice of “significant elements” may appear 
arbitrary, but later the reasons for my selection should become apparent. 



48 


THE HAMADSHA 


Each of the episodes appears, literally, to be concerned with the 
relationship between the saints and some other significant figure. In 
each of them the saint overcomes—or at least obtains the recognition 
of—the other figure, his antagonist. The relationships are of power and 
control. The figures are saints—Bu £ abid Sharqi, Sidi 1-Husayn, Sidi 
Qasim, Sidi Rahal, and Moulay Idriss; intellectual leaders—the canonists 
of Fez; political leaders—Moulay Idriss, Moulay Ismail, £ Abdelhaqq, 
and the king of the Sudan; military leaders—again Moulay Ismail; for¬ 
eigners—the king of the Sudan and his soldiers; women—the blind 
servant woman, Lalla £ Auda, and even £ Aisha Qandisha; servants—Sidi 
Ahmed Dghughi; members of the “spirit realm”— £ A‘isha Qandisha, and 
the angels and jnun who help the saint send oil to Mecca; and the gen¬ 
eral populace—the people of Moulay Idriss. Sidi £ Ali also has control 
over nature. He is able to defy the laws of nature by sending Sidi 
Ahmed instantaneously to Meknes and even to the Sudan, by stopping 
the course of the sun, and by changing soldiers into frogs. Significantly, 
none of the legends indicate Sidi £ Ali’s relationship with God. He is, 
of course, considered, like all saints, to be an intermediary between 
man and God. 

It is suggested that each of these figures represents or is symbolic of 
a significant aspect of the Moroccan’s experiential world, with which 
he must enter into some relation. These relationships seem to be con¬ 
ceived primarily in terms of dominance and submission. Dominance is, 
of course, the ideal, and the saint is seen to be in a dominant position 
at the end of each of the episodes. In this respect he overcomes all of 
an individual’s major social confrontations but two—the economic and 
the sexual—which are not mentioned. The saint lives on alms. By choos¬ 
ing the world of prayer and meditation, he has placed himself above the 
concerns of a daily life which he, nevertheless, can master. 

The Leitmotif of the legends is the passage of baraka. Baraka is the 
essential quality and emanation of the saint. It is rooted in both the 
physiological and social life of the Moroccan. Baraka passes from 
Bu £ abid Sharqi and several other saints to Sidi e Ali, and then from Sidi 
e Ali to Sidi Ahmed. The passage is disjunctive and does not involve 
women. Usually baraka is passed either from father to son (and daugh¬ 
ter) or from a highly endowed individual or place to a visitor or sup¬ 
plicant. In the case of agnatic passage, I have been told that baraka is 
carried in the semen itself. Women cannot pass on baraka. According 
to the Moroccan Arabs, women do not contribute at all to the heredi¬ 
tary background of the child; they are only a receptacle which receives 



THE LEGENDS 


49 


the male seed. This seed is molded into the infant by angels who de¬ 
scend into the womb in early pregnancy. 9 

Baraka that is passed from father to child is and remains contagious. 
Baraka that is passed from an endowed person or place to a supplicant 
can no longer be transferred. The fact that both Sidi c Ali and Sidi 
Ahmed were able to obtain it in its contagious and enduring form from 
saints from whom they were not descended is exceptional. Their means 
of obtaining it were equally unusual. In both cases the acquisition of 
baraka involved the feminization of the saint. 

Sidi ‘Ali, who has an endowment of baraka in his own right, obtains 
the remainder from Sharqi (and other saints). His innate endowment, 
despite the muqaddim’s subsequent retraction, enables him to acquire 
it by deception; he blinds the slave woman and tricks the saint (who 
is, nevertheless, aware of what Sidi c Ali is doing). Whether it is Sidi 
‘Ali’s own baraka or the baraka he acquired when Sharqi touched his 
arm that is responsible for his ability to turn the mills automatically, it 
is noteworthy that hand-mills are normally turned by women. Sidi c Ali 
demonstrates, thereby, his expertise in woman’s work. He blinds the 
slave woman for revealing his “miracle” to the saint; the saint is an¬ 
gered, throws Sidi c Ali out, and falls sick. Sidi c Ali obtains the baraka 
by pulling the pail of vomit away from the blind woman and drink¬ 
ing it. 10 

It is suggested that the oral incorporation of baraka is analogous to 
insemination. 11 Indeed, material collected by free association related 
baraka to semen. It is rumored that some saints and their descendants 
cure sick women by having sexual relations with them; they give them 
their baraka directly. Semen is, according to the Moroccans of the 
Hamadsha’s background, produced by a “vein” in the man’s back 
which extends downward to the lumbar region. This vein converts 


• Women—and not men—are, nevertheless, responsible for barren marriages, and 
must take measures to insure their fertility. A woman may affect a child during 
pregnancy. If she sees a dwarf or deformed individual, her child will be bom with 
the same deformity; if her husband does not cater to her longings for special foods 
such as liver, and she should then scratch herself, her child will have a birthmark 
in the spot corresponding to the place where she scratched. 

w Thc possible Oedipal implications of Sidi c Ali’s blinding the slave woman (the 
mother) for telling Sharqi (the father) of his (the son’s) miracle (baraka, semen) 
will not be considered here. 

n It may also be connected with nursing (milk is said to contain baraka) or 
fantasies of fellatio (which is not normally practiced in Morocco). In all of these 
instances, power is obtained by ingesting bodily secretions. This is a frequent 
fantasy. 




50 


THE HAMADSHA 


blood into semen; blood itself is said to contain baraka and is frequently 
employed in Moroccan cures and magical practices. 

The fact that Sidi £ Ali obtains his baraka by ingesting defilements is 
not as strange as it might first appear. It should be remembered that 
sexual intercourse, for the Moroccan Arab, is an act of pollution. 12 Un¬ 
doubtedly the element of humiliation both on an individual level—man 
made woman—and on a religious level plays a role here as well. The 
saint, at any rate, becomes “pregnant with or by baraka’’ and is grad¬ 
ually recognized as a saint. This occurs during a period of gestation in 
the University of Fez where he remains immobile and spits, 13 or in Sidi 
Qasim’s garden. In both instances the episodes end with his recognition 
as a saint. He alone is able to take the bread from the sultan; he has 
resisted the pomegranate. Sidi c Ali “gives birth to himself’ as a saint. 
It should be noted that during the period of gestation Sidi £ Ali per¬ 
forms manly tasks—meditation, prayer, spitting, gardening—and that 
at the end he does what even ministers and soldiers cannot do. Sidi £ Ali 
becomes a man. It is at this point that he retires to the Zerhoun and 
encounters (gives birth to) Sidi Ahmed, who is instrumental in the 
performance of his miracles and who finally, after bringing the female 
demon ‘Aisha Qandisha to the Zerhoun, inherits Sidi 'Ali’s baraka. 14 

The legend appears to mirror two different developmental sequences. 
On the one hand, it mirrors the process of impregnation, gestation 
(birth), 15 the raising of a son and heir who is able to bring home a 
woman and to have offspring. On the other hand it mirrors the matura¬ 
tion of man from his acquisition of baraka—the hero’s quest for his 
father—through a period of latency—the sojourn in the university or 
in the garden—to the performance of a number of miracles which 
establish his dominant position within society. It culminates in his death 
and the bequeathing of his powers to an heir. The legend of Sidi e Ali 
appears to represent in a more manifest fashion the maturation of a 


“It is curious to note that Sidi c Ali obtains baraka—a theoretically impossible 
feat—by ingesting vomit. Vomit travels the wrong way on a one-way street. 

“Saliva has been equated with semen by psychoanalysts. I have found no evi¬ 
dence for this symbolic equation in my Moroccan material. 

14 Although I do not pursue this line of thought, I should like to point out that 
many of the tasks performed by Sidi c Ali during his period of gestation are of an 
other-worldly nature. In the end he becomes a saint who performs the hadra. 
Such periods occur frequendy in the lives of saints and mystics throughout the 
world. The extent to which the sexual ambivalence expressed in the legend is 
related among Moroccan Arabs to the saintly role, the trance, etc., merits exam¬ 
ination. 

“ Up to this point the sequence of events is also equivalent to Sidi e Ali’s rebirth 
as a saint who can pass on his baraka. 



THE LEGENDS 


51 


man, and in a more disguised fashion the birth cycle of a woman. The 
male maturation process and the female birth cycle are thematically 
intertwined and appear to be combined in the “rebirth” of Sidi c Ali as 
a saint who has contagious and heritable baraka. 

From a psychological point of view, the legend is concerned with 
the problem of male identity. It is a symbolic portrayal of certain sex¬ 
ual conflicts and, if Bettelheim’s (1955) theory of the envious male is 
correct, may indeed be expressive of the male’s desire for female char¬ 
acteristics and potentialities. It may serve as a schema—or partial 
schema—for these conflicts or this envy. The male child must be im¬ 
pregnated—feminized—in order to become a man. The child who 
actively seeks his father’s power (semen, virility, manhood—baraka, 
miraculous power, saintliness) must play a passive role before the 
father. Fenichel (1966:334) notes, in discussing the passive homosexual, 
that the feminine man, who having identified with the mother becomes 
anally fixated, is an analogous mechanism. 

Actually, “feminine” men often have not entirely given up their striving to 
be masculine. Unconsciously they regard their femininity as temporary, as 
a means to an end; they regard the condition of being a masculine man’s 
“feminine” partner as learning the secrets of masculinity from the “master,” 
or depriving him of his secrets. In such cases, the passive submission to the 
father is combined with traits of an old and original (oral) love of the 
father. Every boy loves his father as a model whom he would like to re¬ 
semble; he feels himself the “pupil” who, by temporary passivity, can 
achieve the ability to be active later on. This type of love can be called 
apprentice love; it is always ambivalent because its ultimate aim is to replace 
the master. 

Not only does the legend suggest that at least some of the Hamadsha 
are suffering from problems of male identity, but the parallels between 
the mechanism outlined by Fenichel and the legend of Sidi c Ali suggest 
further a passive homosexual attitude toward the father and an identifi¬ 
cation with the mother (possibly with anal fixation). It is, of course, 
not possible from the analysis of a legend to deduce a psychology; the 
legend may serve only as a clue. 16 

The resolution of a possible Oedipal conflict in terms of passive 
homosexuality is a dead end for the man anxious to have heirs (to 

“I am, at this point, less interested in the psychological implications of the 
legend—the legend as projective material—than in its symbolic-interpretive func¬ 
tion for the individual bom into a society in which the Hamadsha, their cere¬ 
monies, and their beliefs are a given. 




52 


THE HAMADSHA 


obtain and pass on baraka): no children can be produced. Sidi c Ali 
never gives birth to Sidi Ahmed. “Someone came up to Sidi c Ali. His 
name was Sidi Ahmed.” Nor does he bear any children, in legend or in 
fact. The saint must become a man. 17 After a period of latency which 
culminates in a manly act, he is able not only to perform miracles— 
the most notable is his resistance to the beheader, the castrator, Moulay 
Ismail 18 (again, curiously, through ingestion, this time of oil)—but 
also to raise Sidi Ahmed and bequeath him his power. 

From a sociological point of view, the legend is an attempt to resolve 
the conflict between two different endowments of baraka: institution¬ 
alized baraka, which is passed from father to son (and daughter), and 
personal baraka, which is inherent in the individual and is not heritable. 
Personal baraka is in many ways analogous to Weber’s conception of 
charisma (Gerth and Mills 1958:245-252). This problem is of central 
importance in much of Moroccan history. Clifford Geertz (1968:45) 
has written: 

The problem of who has baraka was indeed in some ways the central theo¬ 
logical problem (if that is not too elegant a word for an issue which rarely 
rose above the oral and practical) in classical Morocco. And to it two major 
classes of answers were given: what we may call the miraculous and the 
genealogical. Maraboutism, the possession of baraka, was indexed either by 
wonder-working, a reputation for causing unusual things to occur, or by 
supposed lineal descent from the Prophet. Or, as I say, by both. But though 
the two principles were often, after the seventeenth century perhaps most 
often, invoked together, they were yet separate principles, and in the tension 
between them can be seen reflected much of the dynamics of Moroccan 
culture history. 

Despite Sidi c Ali’s spiritual genealogy ( silsila ), which is not men¬ 
tioned in the legends, it appears that he is endowed with personal, or 
miraculous, baraka, but not with the institutionalized and heritable 

17 Dundes, Leach, and Ozkok (1970) suggest in their analysis of Turkish verbal 
dueling that “the shift from the boy’s female passive role [in homosexuality] to 
the man’s male active role is an intrinsic part of the process of becoming an adult 
male.” Although homosexuality is perhaps not as common in Morocco as in 
Turkey, it is by no means uncommon. There is considerable scorn for the adult 
male who plays the passive role. Adolescent boys who play the passive role are 
teased and are expected when they are older to assume the active role. 

M Freud (1963:109) notes that repudiation of the feminine attitude is “a result 
of the struggle to avoid castration; it regularly finds its most emphatic expression 
in the contrasting fantasy of castrating the father and turning him into a woman.” 
(Cf. Miner and De Vos [1960] for their findings on castration anxiety among 
Algerian peasants.) 



THE LEGENDS 


53 


kind. His trip to Bu'abid Sharqi—and the other saints—is an active at¬ 
tempt to obtain the latter. This is by definition impossible. A man can 
obtain the baraka of a saint, but this is not heritable. A resolution is 
attempted: Sidi c Ali becomes as a woman and is impregnated with the 
saint’s baraka. If, by some miracle, he were able to bear a son and found 
a lineage, the lineage would not be his own since a woman, according 
to the Moroccan Arabs, cannot pass on baraka (or any other traits). 
To become lineage head he must again become man, but then he can¬ 
not father children who will inherit his baraka, since his baraka is not 
of the institutionalized sort. Self-inflicted androgyny has its ideological 
limits. 19 It is an attempted resolution, as L6vi-Strauss (1963b) might 
suggest, between two polar opposites. Another resolution is attempted: 
the master adopts a pupil. 

Before examining the legend of Sidi Ahmed, it should be pointed out 
that the legend also attempts to resolve a contradiction between baraka 
per se and baraka that is a symbol for mystical learning. The former 
can be passed on only agnatically; the latter can be passed on to any 
willing student, despite his origin. 

The most striking difference between Sidi c Ali’s acquisition of baraka 
and Sidi Ahmed’s is that the former actively seeks it and the latter 
passively receives it. The two methods of acquisition are mirror op¬ 
posites. Sidi c Ali, the active seeker, must become a woman and pas¬ 
sively receive it; once he has received it, he becomes a man again. To 
obtain baraka passively, Sidi Ahmed must be a man—this is symbolized 
in the errands he runs for Sidi c Ali, by the fact that his stomach does 
not swell when he drinks ‘Abdelhaqq’s buttermilk, and that he brings 
back a female demon from the Sudan; once he receives the baraka, he 
becomes a woman. He slashes his head. Head-slashing, here, is the sym¬ 
bolic equivalent of castration. That there has been a displacement from 
penis to head is suggested by the fact that ras refers to the glans of the 
penis as well as to the head. 20 This equivalence—penis = head—is not 
infrequently encountered in psychoanalysis, also (Freud 1941). 

The episodes dealing with Sidi Ahmed are much less ambiguous than 
those concerned with Sidi e Ali. (I exclude here the legends relating Sidi 
Ahmed to the canonists of Fez and to Moulay Ismail, which were not 

“The failure of this resolution is masked by certain hints at rebirth. Sidi c Ali 
becomes pregnant with Sharqi’s baraka and gives birth to himself as a saint who is 
capable of passing on his baraka to his children. This theme is not developed. 

"This interpretation is supported by material collected by free association and 
in Rorschach tests. These will be treated in the seauel to this work. I have no 
evidence for an equivalence between blood from head wounds and menstrual 
blood, as Bettelheim (1955) might suggest. 




54 


THE HAMADSHA 


part of the legendary cycle related by Muqaddim Touijer but were 
told to me by the descendants of Sidi Ahmed.) Sidi Ahmed—whether 
the son of Sidi ‘Ali or a stranger—is treated like a servant or slave. This 
is indicative of the subordinate relationship of son to father which is 
characteristic of Arab society. The errands Sidi Ahmed runs for his 
master are tests of his manhood. He must overcome ‘Abdelhaqq, the 
son of Moulay Ismail, the beheader; then he must go to the Sudan to 
fetch the hal. He brings back a female demon, a woman—only to leam 
of his master’s death, according to the most frequently heard version 
of the legend, and his legacy of baraka. The sequence to this point ap¬ 
pears to mirror the maturation of the boy under the tutelage of the 
father. Then there is an odd twist. Once having inherited his father’s 
power, the son castrates himself, at least symbolically, despite the ad¬ 
monitions of his companions. 

“You are lucky to be alive,” they said. “You are lucky to have children. 
The children will carry on your blood and your baraka. You must not be 
angry or sad. You will always have children. They will always replace 
you.” 


At the moment when Sidi Ahmed’s (and Sidi ‘All’s) presumed desire 
to found a lineage carrying on his baraka has been fulfilled, he renders 
himself a woman, and women cannot pass on baraka. A woman is, in a 
sense, like a man with personal baraka; she cannot pass on her traits, 
and he cannot pass on his baraka. 21 

Psychologically, the legend again appears to be concerned with mas¬ 
culine identity. The son has not followed in his father’s footsteps. Al¬ 
though he has obeyed his father’s commands, including the preparation 
of bath-water, which is usually a female task, he has remained a man. 
There are no indications of a passive, homosexual attitude toward the 
father; however, there are hints of his dependence upon the father. 
Each time Sidi Ahmed goes off on a mission, it is Sidi ‘Ali who saves 
him—whether from the hand of Moulay Ismail, after he has killed the 
famed beheader’s son, or from the soldiers of the king of the Sudan, 
after he has stolen the king’s instruments and his woman. At one level 
of analysis, the head-slashing appears to symbolize the son’s inadequacy 
—his lack of independence—once the father has been eliminated; he 
rejects the masculine role, the role of head of a lineage possessed of 

"The passage of baraka from Sidi c Ali to Sidi Ahmed is not as well developed 
as that from Bu'abid Sharqi to Sidi c Ali. The fact that Sidi Ahmed is not the son 
of Sidi c Ali is not played up. 



THE LEGENDS 


55 


baraka. The “masculine protest”—to use Alfred Adler’s expression— 
has been unsuccessful. He renders himself a woman—the sign of frailty, 
weakness, and inferiority—and thereby incapable of passing on ba¬ 
raka. 22 

From a sociological (and an ideological) point of view, the episode 
appears again—but less clearly—to be an attempt to resolve the dilemma 
of the saint who is possessed of personal but not institutionalized baraka 
(and an attempt to resolve the conflict between the passage of baraka 
per se and baraka as a symbol for mystical learning). Sidi Ahmed, who 
is not the real son of Sidi c Ali, plays the role but must depend upon 
his father. When his master dies and bequeaths him his baraka, he is 
incapable of receiving it—he is after all not the son of Sidi c AIi—even 
though he is capable of bearing children. He is compelled to make him¬ 
self feminine—to castrate himself—in order to receive it. Then as a 
woman he cannot pass it on. The dilemma remains unresolved. Once 
again self-inflicted androgyny is not a satisfactory answer. 

The legendary cycle appears to have two principal parts, each of 
which attempts to resolve certain psychological, sociological, and ideo¬ 
logical problems. Psychologically they are concerned with the problem 
of male identity and, if Bettelheim is correct, even with the male envy 
of the female. Specifically they treat of the son’s relationship with the 
father; they reflect alternative attitudes toward the father. The son 
may either play the passive role, as Sidi c Ali did, and receive his father’s 
power; or, as Sidi Ahmed did, he may withstand the temptation to 
submit to the father and find himself inadequate to the power that is 
eventually bequeathed him. Having resisted submission, he has, never¬ 
theless, become dependent. It is suggested that both the passive role and 

"The extreme mourning probably reflects as well the ambivalence that the 
son felt for the father. As Freud (1963:106) notes: “The more ambivalent the 
relation has been, the more lilcely is the grief for the father’s loss to turn into 
melancholia.” 

It is perhaps possible to argue that Sidi Ahmed slashes himself because he feels 
that bringing back 'Aisha Qandisha is somehow responsible for Sidi 'Ali’s death. 
Sidi c Ali did not ask him to bring back 'Aisha; he asked him to bring back the 
hal. ‘Aisha may be seen as a potential wife or as a mother; she is sometimes 
referred to as Sidi 'Ali’s servant. This suggests to the psychoanalytically oriented 
that Sidi Ahmed’s trip to the Sudan to fetch the b*l involved intercourse with his 
mother. The trance state may be symbolic of intercourse. This intercourse may 
have been held responsible for his father’s death. Head-slashing results—if this 
argument is pursued—from the guilt or remorse the son feels for having caused 
his father’s death by sleeping with his mother. 

It may even be argued—and such an argument would help to explain the cura¬ 
tive value of head-slashing—that the mutilation libidinizes the head-penis and 
thereby renders the son (the patient) potent, virile, a man (cf. Bettelheim 1955: 
81-82). 




56 


THE HAMADSHA 


the dependency-inadequacy role of the son are symbolized regardless 
of their genesis in terms of woman. The female is stereotyped as frail, 
weak, inadequate, untrustworthy, and inferior by Moroccan Arabs, and 
as such she represents such self-feelings in the man. The therapeutic 
process will involve, as we shall see, an attempt to realize these feel¬ 
ings symbolically. It is a period of organized regression in which the 
individual symbolically plays out his female identity in order to become 
a man once more. 

Sociologically, the legend attempts to resolve the problem of acquir¬ 
ing baraka which is inheritable from someone from whom one is not 
descended. In broader terms, it is the problem inherent in the Arabs’ 
extreme agnatic principles. The son can acquire only what the father 
possesses; the mother can pass nothing on. 23 Self-inflicted androgyny is 
an attempted but, in the final analysis, unsatisfactory answer to the 
dilemma. Ideologically, the legends attempt again, through the symbol 
of androgyny, to resolve the conflict between institutionalized baraka, 
which can only be passed on agnatically, and baraka as a symbol for 
mystical learning, which can be conveyed to anyone who is willing to 
receive it. 

Since baraka is the curative substance par excellence , it has been 
necessary to follow out these different levels of analysis in order to 
come to some understanding of its logic. At the same time, it has given 
us a means of considering how the problems of baraka transference are 
reflections of possible underlying psychological conflicts. It is sug¬ 
gested that such conflicts would be symbolized and worked through, 
at least for the Hamadsha, in terms of the logic of baraka and the other 
motifs in the legendary cycle. 

a The lack of a well-developed mother figure in the legends is indicative of this. 
As a symbol of weakness, a woman is not able to mediate, ideologically at least, 
the son’s relationship with the father. (Compare Levi-Strauss’ analysis [1956b] of 
the Oedipal myth.) 



PART TWO 



The following four chapters deal with the social and economic organi¬ 
zation of the Hamadsha in three settings: the villages of Beni Rachid 
and Beni Ouarad, the madina of Meknes, and the shantytowns, espe¬ 
cially Sidi Baba, of the same city. My discussion in Chapter Three of 
Beni Rachid and Beni Ouarad is, strictly speaking, concerned less with 
the Hamadsha as members of a religious brotherhood than with the 
descendants of Sidi c Ali and Sidi Ahmed, who may or may not be mem¬ 
bers of a Hamadsha team but who are, by virtue of their birth, inti¬ 
mately related to the Hamadsha brotherhoods. In anthropological 
terms, membership in the %vulad siyyid is ascribed and in the Hamadsha 
brotherhoods achieved. Only those aspects of the life and organization 
of the wulad sayyid which have a bearing on the liamadsha complex 
are covered. 

Attention is then given to the relationship of the saints’ descendants 
both to the cult of their ancestors and to the brotherhoods. Examina¬ 
tion of both of these relationships in the last chapter of this section, on 
social organization, involves an analysis of the passage of baraka and the 
passage of wealth, which complement and reinforce each other. 




4 


The 

Saints’ 

Villages 


The tombs of Sidi ‘Ali ben Hamdush and Sidi Ahmed Dghughi are 
located, respectively, in the two neighboring villages of Beni Rachid 
and Beni Ouarad, on the south face of the Jebel Zerhoun. The Zerhoun 
itself is an irregular massif, covering an area of a little more than a 
hundred square miles and extending from the plains of the Oued 
Sejra, on the south, which separate it from the plateau of Meknes, 
to the Zaggota Pass and the Middle Sebou on the north; it lies west 
of Fez. The massif rises dramatically, like a giant bulwark, out of the 
rich Sais wheat and barley plains, to an altitude of 3,671 feet. Still 
wooded with holm-oak on its crest, it contrasts sharply with the char¬ 
acteristic plains and rolling hills of the area, which are often culti¬ 
vated to their summits, and it is considered by both Moroccans and 
Europeans as one of the most beautiful mountains in the country 
(Boulanger 1966:232; ben Talha 1965:7; Herber 1923). The popula¬ 
tion of the Zerhoun has settled on the periphery of the mountain, 
and its wide crest has remained more or less uninhabited. Both the 
people of the south slope, who are primarily Arabized Sanhaja, and 
the people of the north slope, who tend to be of Rifian origin, live in 
clay-block (or sometimes stone) houses which are covered with ter¬ 
raced roofs of beaten earth. 

The Zerhoun is said to produce the best olives in Morocco, and 
the sides of the mountain, particularly on the richer, south side, are 
covered with olive trees wherever there is sufficient soil for them to 
take root. The cellars of the larger houses in many of the villages— 
there are six principal villages on the south face alone—are equipped 




62 


THE HAMADSHA 


with oil presses resembling the presses found in the nearby Roman 
ruins of Volubilis, locally called the Pharaoh’s Palace. The trees may 
be individually or collectively owned, worked by their owner or one 
of his hired hands, or rented out by individual agreement or to the 
best bidder in a village auction. Although occasional fig, carob, citrus, 
and other fruit trees can be found in the vicinity of the villages, the 
olive is the mountain’s principal economic resource. (Prickly pears, 
which grow in the drier, rockier gorges and gullies, are also sold.) 
Many of the villagers, however, also own, or at least work, fields in 
the plains; they store their wheat and barley in underground granaries. 
There are also kitchen gardens, and many families have a few sheep, 
or even a cow, grazed by a son, grandson, or hired hand. The villages 
on the south face of the Zerhoun, like Beni Rachid and Beni Ouarad, 
are economically oriented to Meknes; those on the north face to 
Moulay Idriss, and to a lesser extent Fez. 

Although the Zerhoun, located as it is between the imperial cities 
of Meknes and Fez, has long been part of the blad l-makhzan , or 
region under central government control and taxation, it has tradi¬ 
tionally been a refuge for victims of war and famine, such as the Rifian 
Berbers who settled on the north face. The refugees were attracted to 
the mountain not only because of its strong defensive position but 
also because of its agricultural potential and its opportunities for work. 
They were drawn, too, by its sanctity, for the Zerhoun, accommodat¬ 
ing the tombs not only of Sidi c Ali and Sidi Ahmed but also of Moroc¬ 
co’s “national saint,” Moulay Idriss, and many lesser saints of merely 
local importance, is considered a sacred mountain, and its inhabitants 
are famed for and proud of their piety. 1 These migrations to the 
mountain have produced a heterogeneous population, both with 
respect to the Zerhoun as a whole and within many of the villages 
themselves. This is especially true of Beni Rachid and Beni Ouarad. 

In the years immediately preceding the arrival of the French in 
Morocco in 1912, the Zerhoun was divided into two administrative 
units, one for the north, centered in Moulay Idriss, and the other for 
the south, centered in El Merhasiyne. Each district was headed by a 

1 They are proud of the fact that there are no Jews or Christians on the Zerhoun 
and that, despite the unusual number of pilgrimage centers, there are no prosti¬ 
tutes. Prostitutes have tended to gravitate to active pilgrimage centers like Moulay 
Brahim and Sidi Rahal near Marrakech, and the baths of Moulay Yacoub near 
Fez and Sidi Sliman Mula 1-Kifan near Meknes. It is said that the authorities 
will not touch them in the sacred area of the sanctuary. Ironically, the two baths 
are visited by those suffering from venereal diseases. 



THE SAINTS’ VILLAGES 


63 


caid, or local administrator, who was responsible to the central gov¬ 
ernment, and each village was headed by a muqaddim. The muqaddims 
of Beni Rachid and Beni Ouarad were responsible to the caid of El 
Merhasiyne; they were members of the saintly families but were not 
the leaders (mizwars) of these families. From what I have been able 
to gather from the older peasants, the traditional Moroccan Arab 
council of elders (j?na c ) was of little importance in the two saintly 
villages even in the prc-Protectorate days, and village life was dom¬ 
inated, as it is today, by the leaders of the saints’ children. 

The French maintained this arrangement for some time after their 
arrival, but they eventually simplified it by placing a single muqaddim 
at the head of not one but three villages. Significantly, the muqaddim 
charged with Beni Rachid, himself a resident of the village and a 
descendant of Sidi c Ali as well as brother-in-law to the caid of El 
Merhasiyne, was not charged with the administration of Sidi Ahmed’s 
village of Beni Ouarad but with two other neighboring villages. This 
administrative separation probably resulted from the hostility and 
competition which has existed between the two villages for many 
years and which is, as we will see, dramatically portrayed during the 
annual pilgrimages to the saints’ tombs. 

Since Moroccan independence in 1956, a single caid has been re¬ 
sponsible for the entire Zerhoun; his headquarters are in the town of 
Moulay Idriss. Moulay Idriss, however, falls under the separate charge 
of a pasha rather than under the jurisdiction of the Zerhoun caid. 
Beni Rachid and Beni Ouarad are under the more direct control of a 
sheikh, subordinate to the caid, who lives in the village of Hamraoua. 
(The former caid of El Merhasiyne lives in exile in France.) The 
villages still have different muqaddims, but the muqaddim of Beni 
Rachid is no longer a resident of the village or a descendant of its saint. 

Beni Rachid and Beni Ouarad 

Both Beni Rachid and Beni Ouarad, like the other villages on the 
south slope of the Zerhoun, are comparatively wealthy by rural Mo¬ 
roccan standards. Beni Rachid, which is much richer than Beni Ouarad, 
is located about a mile farther down the El Merhasiyne-Moulay Idriss 
road and is separated from Beni Ouarad by the hamlet of Qelaa, 
where some of Sidi ‘Ali’s descendants live. A narrow, winding path, 
frequented by pilgrims to the saints, joins the three villages. The houses 
of Beni Rachid are clustered on top of a rocky mamelon and are often 




64 


THE HAM ADS HA 


supported by the walls of an old kasba (fortress), now in ruins, which 
is said to have belonged to one of Moulay Ismail’s qaids, or pashas, 
called bel Shaqur. The villagers maintain that bel Shaqur, like his sultan, 
was notorious for his extreme cruelty and was finally flogged to death 
by the peasants of the nearby village of Beni Jennad after demanding 
the right of the first night from their virgins. 2 His kasba, they insist, 
was at least twenty kilometers in circumference; and although this is 
an obvious exaggeration, the kasba must have been an impressively 
large structure, since many of the houses of Beni Rachid lie within its 
confines. This gives the village a very tight, concentric look which 
contrasts with the narrow, elongated appearance of Beni Ouarad. 

Sidi 'Ali’s mausoleum, just above the basin and baths of the 'Ayn 
Kabir, runs lengthwise on a sort of land bridge that joins the mamelon 
of the village on the west with the Qelaa ridge on the east. It is one of 
the largest mausoleums in the country, measuring some 150 feet in 
length. The tomb, which forms an almost perfect square about 45 feet 
on each side, is located at the western end of the mausoleum complex 
and is covered by a hexagonal roof of glazed green tile. (Most of the 
qubbas in which Moroccan saints are buried are square with domed 
roofs.) An open rectangular court, about 110 feet by 45 feet, extends 
eastward from the tomb. One enters the court from the south and the 
tomb from the courtyard. Along the interior walls of the court there 
are little cubicles built by Sidi 'Ali’s children for pilgrims who wish to 
spend the night, but these cubicles are not very popular. They are 
infested with fleas, ticks, and lice, and the pilgrims prefer to battle the 
insects in the sanctity of the tomb itself, where they can garner more 
of the saint’s baraka. Sidi 'Ali lies parallel to the western wall of the 
mausoleum. His face is turned toward Mecca, and he is covered with 
a catafalque ( darbuz ) which consists of a simple wooden frame, cov¬ 
ered with several layers of red, yellow, and white cloth ( ghota; kaswa; 
or, if made only of cotton, izar). An alms box stands next to the door 
of the tomb room. At the southeastern comer of the complex is a yel¬ 
low and white minaret, next to a more traditional qubba in which Sidi 
'Ali’s disciple—some say his son and others his brother from whom 
the children of the saint are, in fact, descended—Sidi 1-Hafyan, the 
barefoot saint, is buried. The entire complex, which is usually referred 
to as the siyyid , has been under restoration for the past few years. 
The courtyard has been added, and the buildings have been plastered 
and whitewashed. The mizwar, who is responsible for the building, 

8 There are no legends relating bel Shaqur to either Hamadsha saint. 



THE SAINTS’ VILLAGES 


65 


plans to tile the court, put a fountain in the middle, and illuminate the 
whole complex when electricity finally comes to Beni Rachid. 

Although there are several other saints buried in Beni Rachid—the 
most notable is Sidi Musa, from whom many of the villagers claim de¬ 
scent—they do not play a significant role in the cult to Sidi c Ali. 3 The 
c Ayn Kabir and a grotto to c A'isha Qandisha are, however, important 
pilgrim stops. The waters of the c Ayn Kabir flow from the rocky wall 
of the land bridge on which Sidi c Ali’s tomb is located. They are con¬ 
sidered to contain much baraka, and are valued for their therapeutic 
effects. A few years ago the descendants of Sidi c Ali attempted to build 
a basin for the water but, as the basin cracked open every time they 
completed it, they decided that Sidi c Ali did not wish to have his 
sacred waters confined, and they abandoned the project. The water 
now drips down a stairway to the three-roomed, clay-block bathhouse 
which the saint’s descendants built at the same time they tried to build 
the basin. They charge pilgrims an admission of 50 francs (or 100 
francs during the musem) a bath. 4 

The grotto ( hufra ) to c Aisha Qandisha lies a few hundred yards 
east of the village proper. It is banked on one side by the root system 
of an enormous fig tree, which towers over it and is considered ex¬ 
tremely powerful by local peasants who graft its branches onto their 
own young trees to make them fruitful. The grotto itself consists of 
two small hollows. In the larger outer hollow, the roots of the fig tree 
are tied with rags and ribbons and other bits of cloth which have been 
left by pilgrim women as a sign of a vow ( c ar ) to sacrifice something, 
usually a black chicken, to c Aisha Qandisha if she grants them their 
wishes. The inner cavity, where an underground stream trickles to the 
surface, is dark and muddy. It is here that c Ai‘sha is said to reside. 5 

The mausoleum of Sidi Ahmed, which is located at the end of the 
main road that separates the larger houses of the wealthy villagers on 
the south from the smaller houses of the poor on the north, is much 
smaller than Sidi e Ali’s tomb and, unlike it, is structurally integrated 
with the houses of Beni Ouarad. Not only does the house of the 
former mizwar actually share a wall with the mausoleum, but the 
square in front of the tomb is the scene of most village activity. Al¬ 
though Sidi Ahmed’s descendants claim that their ancestor’s tomb is 

8 There are no legends relating them to either Sidi c Ali or Sidi Ahmed. 

4 There are 100 francs to the dirham. A dirham was worth 20 American cents at 
the time of my investigation. 

*1 have heard ‘ATsha’s grotto called the fish spring ( l- c ayn l-hut) by women. 
It should be noted that c ayn may mean both a spring and an eye. Fish are asso¬ 
ciated with other sanctuaries in Morocco. 



66 


THE HAMADSHA 


also undergoing renovation, there are no signs of work, with the pos¬ 
sible exception of a few dabs of whitewash around the “keyhole arch” 
door. The tomb itself, which can only be reached through the court¬ 
yard, is located to the northwest of the courtyard and is oriented 
from northeast to southwest. In the corner of the courtyard is a small 
hole, filled with muddy earth, which is called ‘Aisha’s pit ( hufra) 
because e Ai'sha Qandisha is said to appear there from time to time, 
especially when ecstatic dances are performed in front of the tomb. 
Sidi Ahmed’s descendants are very sensitive about this pit, because it 
does not compare either in size or lucrativeness to ‘Aisha’s grotto at 
Beni Rachid. They believe that since their ancestor brought ‘Ai'sha to 
the Zerhoun, they should obtain a share of the profits from the grotto, 
but the descendants of Sidi c Ali have never agreed with them. 

Population and Social Organization 

The two villages of Beni Rachid and Beni Ouarad are about the same 
size. Beni Rachid, according to the census of 1960, has a population of 
500 (89 households), of whom about 300 claim descent from Sidi e Ali. 
Beni Ouarad, according to the same census, has 604 inhabitants (140 
households), 200 said to be the children of Sidi Ahmed. Although the 
saints’ descendants account for only 60% of the population of Beni 
Rachid and 30% of Beni Ouarad, they dominate village life both eco¬ 
nomically and politically and are treated with a mixture of respect 
and resentment by the other villagers. There are almost no marriages 
between the descendants of the two saints. Of 146 marriages recorded 
for the children of Sidi Ahmed, only two took place with inhabitants 
of Beni Rachid, and only one of these with a descendant of Sidi c Ali. 
Although over 50% of the marriages recorded were with spouses from 
outside the village, and about 23% with inhabitants of other villages on 
the Zerhoun,® it appears that there is a great reluctance to form unions 
not only with the descendants of Sidi e Ali but with any inhabitants of 
Beni Rachid. Despite the traditional Arab preference for parallel- 
cousin marriage, only two parallel-cousin marriages were recorded 
for Sidi Ahmed’s children. And in spite of the obvious economic ad¬ 
vantages of marrying a descendant of the saint—the saint’s children 
receive a share of the proceeds from the saint’s tomb and of the gifts 
of the brotherhood—only 20% of the marriages in the last four genera- 

• It should be noted that no marriages with inhabitants of Qelaa and Beni Jennad, 
the next closest villages, were recorded. Qelaa is dominated by the descendants of 
Sidi c Ali. 



THE SAINTS’ VILLAGES 


67 


tions were endogamous. 7 Although my figures are not complete for 
Sidi c AIi’s children, who were quite resistant as informants, a similar 
pattern is discernible in even my fragmentary data. There are consider¬ 
able differences in wealth and political influence among the descend¬ 
ants of each saint, but the children of Sidi ‘Ali appear, on the whole, to 
be wealthier than those of Sidi Ahmed. 

The descendants of Sidi c Ali and Sidi Ahmed follow, at least ideally, 
the traditional model of segmentary society characteristic of the Arab 
Bedouin (Peters 1960; Evans-Pritchard 1949:29-61), and they are di¬ 
vided into agnatic lineages whose members orient themselves to a 
significant ancestor and refer to themselves as the children {wulad) 
of that ancestor. The particular ancestor chosen depends on the par¬ 
ticular segment of the group the speaker wishes to call on and on 
what is demanded by the particular circumstances at hand. Since 
the descendants of both saints live, and have lived for hundreds of 
years, under central government authority, the corporate strength of 
the descent groups has been greatly—but not completely—reduced 
to the field of domestic relations (cf. Fortes 1953). The division of 
proceeds from the saints’ tombs ( futuhat ), the religious brotherhoods, 
and the saints’ properties ( Ipabus ), as well as the administrative func¬ 
tion of the mizwars as the heads of each of these saintly lineages and 
the brotherhoods, has preserved the corporate structure of the groups 
to an extent which is not apparent among the other inhabitants of 
the Zerhoun. 8 

The most distant ancestor, encompassing the largest descent group 
capable of joint action, to which the descendants of either saint make 
reference is either Sidi c Ali or Sidi Ahmed. (Both descent groups 
appear to have no affiliation with any larger clan or tribal groups.) 

7 It is theoretically possible to argue that the high frequency of exogamous 
marriages resulted from the saintly children’s desire to increase their political 
support. Given the complexities of political power within the saintly lineages and 
within the village itself, this seems unlikely. A better argument is that nonsaintly 
families are willing to accept smaller bride-prices and give larger ones for the 
prestige of marrying into a saintly family. 

•“A group may be spoken of as ‘corporate’ when it possesses any one of a 
certain number of characters: if its members, or its adult male members, or a 
considerable proportion of them, come together occasionally to carry out some 
collective action—for example, the performance of rites; if it has a chief or council 
who are regarded as acting as the representative of the group as a whole; if it 

E ossesses or controls property which is collective, as when a clan or lineage is a 
ind-owning group” (Radcliff e-Brown 1962:41). The descendants of the Hamad- 
sha saints score on each count: they come together during the annual pilgrimage 
and for the division of proceeds; they have a chief, the mizwar; and they hold 
habus lands in common. 



68 


THE HAMADSHA 


Generally, they refer to themselves and are referred to as 'wulad 
siyyid, the children of the saint, when they wish to distinguish them¬ 
selves from, say, the other villagers, or ivulad Sidi t Ali or wulad Sidi 
Atoned when they wish to distinguish themselves from the descendants 
of another saint. Since every member of the wulad siyyid is at least 
theoretically capable of tracing his genealogical connection with every 
other member of the group from a known ancestor, the wulad sayyid 
is then, in anthropological terms, a maximal lineage (cf. Radcliffe- 
Brown 1962:39). The mizwar is the leader of the wulad siyyid, as 
well as of smaller lineage segments within it; and the permanence of 
his position, which is chartered by the “extra-familial hand” of the 
King as representative of the central government and the Moroccan 
nation, tends to give the group a more stable character than is usually 
found in “pure” segmentary systems. Even among the smaller seg¬ 
ments of the wulad siyyid, there is a similar tendency toward per¬ 
manency. The particular segments which play a decisive role in the 
lives of the saint’s descendants are those concerned with the division 
of proceeds from the saint and the brotherhood. It is these that have 
more or less congealed. 

All men, women, and children who are agnatically descended from 
either Sidi *Ali or Sidi Ahmed are eligible for an equal share in the 
proceeds from the saint’s complex and the associated brotherhood. 
The proceeds from the former include the alms, candles, sacrificial 
animals, and other gifts left by pilgrims at the saint’s tomb or for 
‘Aisha Qandisha, as well as the fees from the baths in the case of Sidi 
‘Ali’s descendants. Profits from the habus lands—lands donated to the 
saint by pious Muslims who are anxious to obtain by their good deeds 
much baraka and, thereby, easy entrance to heaven—are also divided. 
The teams associated with each saint are required to bring large gifts— 
theoretically all that they have received in the form of gifts and fees 
throughout the year—to their mizwar at the time of the musem. 
These, too, are traditionally divided among the saint’s descendants.® 

# A distinction is usually made between alms ( futuhat ) left by pilgrims and the 
revenues of the sanctuary’s habus properties. The former are most often divided 
among the saint’s descendants; their manner of distribution lies ultimately in the 
hands of the King. Revenue from the habus properties is supervised by an official 
overseer {nadir). Conflicts concerning the distribution of alms were to be decided 
in the last instance by a decree from the King; those concerning habus revenues 
by the religious courts (Anonymous 1927?). Such legal niceties were not generally 
recognized by the wulad sayyid at either Beni Rachid or Beni Ouarad. They 
claimed that not only alms but habus revenues and revenues from the zawiyas 
and teams were traditionally evenly divided among them, and ought still to be so 
divided. 



THE SAINTS’ VILLAGES 


69 


Among the descendants of Sidi £ Ali, there are four major lineages, 
each claiming descent from one of Sidi ‘Ali’s four supposed sons, 
which have traditionally played a role in this division. A single respected 
elder from each was always present at the division, which took place 
in the confines ( horm ) of the saint’s tomb. He received the shares 
set aside for his lineage and distributed them. Although the presence 
of the elders in the sanctuary assured each lineage of its proper share 
—barring of course collusion, which was unlikely in the sacred pre¬ 
cincts {horm) of the mausoleum—there was no institutionalized safe¬ 
guard against cheating individuals, or individual family heads, within 
each lineage, and this gave rise to a great deal more suspicion and 
mistrust among the saint’s descendants than probably occurred in 
most Moroccan Arab villages. 10 

To what extent this is a realistic portrayal of the division at Beni 
Rachid is impossible to determine, for the system has not been opera¬ 
tive for several years. Judging from the complexities of the lineage 
structure among the descendants of Sidi Ahmed, about which my data 
are much more complete, it seems unlikely that the lineage structure 
was this simple or that it accounted, as I have been told again and 
again, for all of the children of Sidi c Ali. The mizwar, assured of his 
position both by his wealth and by outside (government) authority, 
has not divided the proceeds from the saint’s tomb in several years, 
but has claimed to have used them for the restoration of the mauso¬ 
leum. This has given rise to even more suspicion and resentment 
among Sidi c Ali’s children, who do not feel that they were properly 
consulted and do not believe that the progress of restoration is in any 
way commensurate with the expenses claimed. Rumor has it that the 
mizwar pockets most of the proceeds and pays out the rest to the 
more influential of his relatives. Thus the major lineages, which had 
traditionally played an important role in the division, have lost their 
principal function. Still, when asked, the villagers assured me that the 
division took place in the prescribed way until very recently and 
that it will again take place as soon as the renovation is completed (or 

there is a change of mizwar). 

The lineage organization of the descendants of Sidi Ahmed, which 
is diagrammed in Figure 1, is more complex. The lineages do not 
play as decisive a role in the division of proceeds as is alleged for 
the major lineages of the wulad Sidi c Ali. Division is organized here 

10 Such suspicion and mistrust is probably characteristic of the descendants of 
other saints as well. It was, for example, very noticeable among the “children” of 
Aloulay Brahim. 



70 


THE HAMADSHA 



^ SI D1 AHMED DGHUGHI ® LINEAGE TRACED FROM WIFE 

A FOUNDER OF LINEAGE «- RECEIVES NO PROCEEDS FROM SAINT 

Figure 1. The lineages of the wulad Sidi Ahmed Dghughi 

quite differently, and operates on chance. Each of the agnatic descend¬ 
ants of Sidi Ahmed—man, woman, and child—who qualifies for a 
share is in charge of the sanctuary for a day and receives all of the 
proceeds of the day with the exception of a bull, which is divided 
equally among the saint’s descendants. 11 Rotation stops for the very 
active month preceding and including the musem; all of the proceeds 
for this month go to the mizwar, who uses them for preparing for and 
entertaining at the pilgrimage. The role of the lineages is limited then 
to the division of gifts from the brotherhoods (and whatever remains 
from the month of the museum). This division takes place not in the 
mausoleum but in the mizwar’s house. 

Among the ten lineages we find that three do not receive a share 
because they are said not to be “true” children of Sidi Ahmed. The 
members of two of these are no longer residents at Beni Ouarad. Of 
the remaining seven lineages, one is virtually defunct, having one last 
member, an old woman, who receives her share but does not partici¬ 
pate in the division. Elders from only four of the other six lineages 

“There are no special rules for butchering. The meat is distributed without 
reference to cut. 



THE SAINTS’ VILLAGES 


71 


have supervised the sharing: wulad Hajj Qasim, wulad ‘Ammariyya, 
wulad Fasiyya (of whom the present mizwar is a member), and wulad 
khai Driss (who may have represented wulad khai Muhammud as 
well). The wulad Muhammud ben Driss, one of the largest, has not 
been represented recently but has received a share. It is from this 
lineage that the mizwars came before the position was preempted for 
his son by the father of the present mizwar, who was the wealthiest, 
most influential man in the village. 

It must be emphasized that representation does not follow a hard 
and fast rule but depends more on local politics. The wealthiest and 
most influential elders, regardless of lineage affiliation, are chosen. 
There is a tendency for these elders to represent different lineages— 
which may be structurally justified by the fact that the largest possible 
number of descendants are, thereby, represented and appeased—but this 
is not always the case. The division the year before I made this study 
was supervised by two elders from the wulad Fasiyya and by repre¬ 
sentatives of the wulad Hajj Qasim and the wulad ‘Ammariyya. It 
appears that among the descendants of Sidi Ahmed, the lineage struc¬ 
ture does not operate in terms of balancing opposing groups of equal 
size and status, as such segmentary systems are said classically to 
operate, but in terms of local political considerations. This probably 
results from the fact that the authority of the mizwar rests on his 
wealth and his influence with the outside world, and it has tended to 
render the structural balance of power ineffective, if indeed it ever 
was effective in a world as open to outside influence as that of Beni 
Ouarad. 

The Mizwar 

The mizwar of each of the saint’s families is in complete control of 
the affairs of the saint’s complex and is considered to have a greater 
endowment of baraka than the other of the saint’s descendants, al¬ 
though the descendants themselves and the most pious of the fol¬ 
lowers do not make this assertion straightforwardly. Not only is the 
mizwar charged with the care and maintenance of the saint’s tomb 
and associated cults, but he is the leader of all of the teams that claim 
to be followers of his ancestor. He also administers all habus lands. 
Although his powers are officially limited to his descent group, where 
he serves as an arbiter in the case of disputes, he is in fact in control 
of the village because of his position as mizwar, his wealth, and his 
influence with the government. 



72 


THE HAMADSHA 


The mizwar’s political powers outside his maximal lineage and the 
village are restricted. Although he is charged with the administration 
of a network of lodges found in all of the principal cities of northern 
Morocco and in many rural areas as well, he does not have under his 
control a politically significant organization from either a formal (in¬ 
stitutionalized) or informal point of view. The adepts were, and still 
are, without political influence. Moreover, the saintly families, un¬ 
like those of other similar complexes like the Ihansalen of the central 
High Atlas (Gellner 1963), have played almost no institutionalized 
political role in the surrounding tribal and village groups. This may 
be accounted for on two grounds. First, the descendants of Sidi c Ali 
and Sidi Ahmed have lived in the center of the blad 1-makhzan and 
have been subjected to the central authority of the sultanate from the 
time of the death of their saintly ancestors. Second, the tombs of Sidi 
‘Ali and Sidi Ahmed are located within the sphere of influence of 
the greater saint, Moulay Idriss. Thus, even asylum in the sacred con¬ 
fines of the mausoleum—asylum could be sought, as in the Christian 
churches of Europe, in many of the saints’ tombs of Morocco 
(LeTourneau 1949:600)—was seldom preferred to that of Moulay 
Idriss, whose descendants had much more influence with the central 
government than those of either Sidi *Ali or Sidi Ahmed. 

There are no explicit rules for succession to the “mizwarship.” All 
that is required is a testimonial letter signed by twelve of the saint’s 
descendants, which is then presented to the government for ultimate 
approval by the King. Among the descendants of Sidi ‘Ali, three elders 
from each of the four major lineages were said to have approved the 
choice of mizwar; among the descendants of Sidi Ahmed, the situa¬ 
tion was more complicated. The mizwarship has tended to pass from 
father to son in both groups, however, and any break in this pattern 
is considered by both the saint’s descendants and followers and by 
other villagers as a break with tradition ( qa'ida) which will incur the 
saint’s disapproval. When Taik, the former mizwar of Sidi c Ali, died, 
for example, he left a son and several grandsons. All of them were 
eligible for the mizwarship theoretically, but the present mizwar, 
Taik’s younger brother, who was wealthier than Taik and had more 
influence with the government than his brother’s sons and grandsons, 
obtained the position for himself by paying off twelve elders to witness 
his appointment. He was able then to obtain—at a price, it is usually 
claimed—a letter of charter from the King which insured his appoint¬ 
ment. The fact that despite three marriages he had only a single son 
who was bom before he became mizwar was taken by the villagers 



THE SAINTS’ VILLAGES 


73 


as a sure sign of Sidi c Ali’s disapproval, and it was not until he “ar¬ 
ranged matters” by marrying his son to Taik’s granddaughter that 
Sidi c Ali’s “approval” was gained. The mizwar then married still an¬ 
other woman and had six children by her. 

The Endowment of Baraka 

All of the men, women, and children who are descended through 
the male line from either Sidi c Ali or Sidi Ahmed are endowed from 
birth with their ancestor’s baraka. It is necessary once again to dis¬ 
tinguish between the two different endowments of baraka: institu¬ 
tionalized and personal baraka. The former results from the status of 
the individual, and the latter from his personal merit. As agnatic 
descendants of a saint, all of the children of the saint are institutionally 
endowed with baraka, without regard to their personal merit. It is 
asserted, however, that as descendants they must necessarily be en¬ 
dowed with the qualities that are said to be in keeping with the pos¬ 
session of baraka. This baraka can never be lost. The second type of 
baraka, the personally achieved, depends not upon an individual’s 
heritage but upon his character, his piety and spirituality, his moral 
fiber and his therapeutic gifts—in short, upon that “vividness” of 
which Geertz writes. 

Although all of the agnatic descendants of the saints are possessed 
of baraka—their institutional endowment—some are considered to 
have more baraka than others. This greater endowment may also be 
institutionalized, as in the case of the mizwar, or achieved, as in the 
case of certain elders who by virtue of their personal qualities are 
said to have a lot of baraka. (Old people are generally said to have 
baraka.) The latter, since they are approachable, are often sought 
after by pilgrims for cures. The mizwars themselves usually keep a 
certain distance from the run-of-the-mill pilgrims, in keeping with 
their highly esteemed position. This distance not only safeguards their 
authority, but it also preserves their institutionalized endowment of 
baraka from any dilution by their lack of personal baraka. 

Endowed from birth with their ancestor’s baraka, the descendants 
of the saint have no need for any special ceremonies by which their 
endowment is made manifest or confirmed. Their attitude toward 
their ancestor is not qualitatively different from that of other Moroc¬ 
cans to the saint, except insofar as they regard him as an economic 
boon. They follow all of the religious practices of the average Moroc¬ 
can of the Zerhoun (cf. ben Talha 1965), but invite the local team to 



74 


THE HAMADSHA 


perform at their name-day celebrations, circumcisions, and other life- 
crisis ceremonies more often than do the other inhabitants of the 
massif. They recite communal prayers in the mausoleum on the night 
of the twenty-seventh of Ramadan (Lilat l-qadar). They follow essen¬ 
tially the same procedures as other pilgrims when they visit their an¬ 
cestor’s tomb, but they do not give any alms. 

Some of the saint’s descendants belong to the local Hamadsha team, 
whose members are recruited primarily from the other villagers. 
Others, like the mizwars and the wealthier descendants, never par¬ 
ticipate in any of the team’s dances and treat their activities with a 
mixture of scorn and respect. The teams of both villages, which are 
very loosely organized and poorly led, are rife with jealousy and fac¬ 
tionalism. Members who are descended from the saint believe they 
should receive all of the proceeds; the other members believe that the 
proceeds should be shared equally; and the saint’s descendants who 
are not team members think that all of the proceeds should be placed 
in the alms box and shared equally among them. Although some of the 
children of the saint do fall into trance and slash their heads, they 
categorically deny—and so do their pious followers—that this has 
ever occurred. They cite a proverb which may be roughly translated: 
“If a child of the saint falls into ecstasy, the followers of the saint will 
fall out of ecstasy.” This proverb not only emphasizes the social dif¬ 
ferentiation between the children of the saint and the followers, but 
it also suggests that there may in fact be a qualitative difference be¬ 
tween the children’s endowment of baraka, which does not enable 
them to trance and slash, and the followers’ endowment, which is the 
enabling factor. 



5 


The 

Lodges of 
Meknes 


Within the city of Meknes the Hamadsha fall into two principal 
categories: those who are associated with one of the two lodges of 
the old town, or madina , and those who are affiliated with teams that 
have no special meeting place. Although the majority of teams are 
found in the bidonvilles that surround the city, there are a number 
of madina inhabitants, usually recent immigrants to the city like their 
confreres in the shantytowns, who are affiliated with them. This chap¬ 
ter deals with the Hamadsha who are associated with the two lodges, 
or zawiyas, of Meknes. They may be divided into the adepts ( foqra ), 
who are members of the lodges; the devotees (muhibbin)\ and the 
ghiyyata, or oboe players. The members of one of these zawiyas are 
followers of Sidi c Ali and are called c Allaliyyin; the members of the 
other are followers of Sidi Ahmed and are called Dghughiyyin. 1 

The City of Meknes 

The city of Meknes is set between the Zerhoun massif on the north 
and plateaus on the south which extend down to the Middle Atlas and 
are populated primarily by Berbers. It lies on the great caravan route, 
the tariq s-suitan, that joined Algeria and other Islamic lands to the 

x The distinction between the zawiyas whose members follow one or the other 
saint does not exist in all Moroccan cities. In some cities there is only one zawiya, 
and followers of both saints attend it; in other cities, like Marrakech, membership 
in the two zawiyas is mixed. The two lodges there are linked to rival quarters 
rather than to any one saint. Even in Meknes, many of the adepts of one lodge 
have transferred to the other. 



76 


THE HAMADSHA 


east with the rich Atlantic plains of northern Morocco. These plains 
had been settled by Hilalian Arabs in the eleventh and twelfth cen¬ 
turies, and Meknes may be considered a sort of border town between 
Arab and Berber worlds. It is located about 35 miles from the imperial 
city of Fez, and with the exception of the long reign of Moulay Ismail 
(1672-1727), when the city was in its prime, it has been overshadowed 
both intellectually and politically by the great university city. 

Originally a scries of small villages, Meknes was first consolidated 
by the Almoravid sultan Yussef ben Tashfin in the second half of the 
eleventh century; but it was not until Moulay Ismail transferred his 
capital to Meknes in the seventeenth century that the city became 
internationally famous. Moulay Ismail, builder with grandiose aspira¬ 
tions, was determined to have his city outshine the cities of Fez and 
Marrakech, both of which he despised because of their resistance to 
his power. He personally supervised the work of about 2,000 Christian 
captives, 30,000 common criminals, and countless black African slaves 
who built his palaces, gardens, military quarters, and kasbas, and the 
more than fifteen miles of walls which surrounded the city (Julien 
1966(II):237). The eighteenth-century English traveler John Windus 
(1725:115-116) has written: 

The Emperor is wonderfully addicted to Building; yet it is a question 
whether he is more addicted to that, or pulling down, for they say if all his 
Buildings were now standing, by a moderate computation, they would 
reach to Fez, twelve Leagues off; And those who have been near him since 
the beginning of his Reign, have observed him eternally building and pull¬ 
ing down, shutting up Doors and breaking out new ones in the Walls. But 
he tells them this is done to occupy his People; for says he, if I have a Bag 
full of Ratts, unless I keep that Bag stirring they will eat their way through; 
but he does not design to give them time. 

It must be remembered that it was during the reign of Moulay Ismail 
that Sidi c Ali lived and the Hamadsha first appeared. After the Sultan’s 
death, Meknes declined rapidly; but it did retain its general plan until 
the French arrival in the beginning of this century. 

The French made Meknes their principal military garrison, and the 
city soon developed into an agricultural center. The appearance of 
garages, retailers in trucks and farming equipment, canning and bot¬ 
tling factories, transportation and trucking services, pharmacies and 
moving picture theaters, had, however, only a gradual influence on 
the madina. Certainly the greater demand for madina products, both 
by Europeans and Aloroccans, had its effect; but the tanners and cop- 



THE LODGES OF M ERNES 


77 


persmiths, the enamelcrs specializing in the grey-black Meknes enamel- 
ware, the tile- and brick-makers, and the other artisans, continued their 
age-old manufacturing practices and went about their business as they 
always had. Although the French changed the bureaucratic structure 
of the city, this too had but little effect upon the average madina Arab, 
who retained his traditional dislike of dealings with the government. 
Even today, more than fifty years since the French first arrived in 
Morocco, madina life in Meknes is much as it was in pre-Protectorate 
days. 

The contrast between these two economies and two life styles is 
illustrated by the contrast between the madina on the northern ridge 
of the Boufkranc and the new quarter, built by the French on the 
site of Moulay Ismail’s olive groves on the southern ridge. The Ham- 
riya, as the new quarter came to be known, followed the plan of the 
traditional French city, with broad, tree-lined avenues, carrefours, and 
modem, windowed houses and apartments. The madina followed the 
plan of the Muslim city, with its narrow, winding streets and twisted 
alleys, its specialized markets, and its ethnically-oriented residential 
quarters with windowless houses, all radiating out from the Friday 
mosque and central market (von Grunebaum 1955:154 et seq.). In 
the case of Meknes, the streets have been laid out according to the 
qibla, the orientation toward Mecca, or to its perpendicular (Planhol 
1959:17). Moulay IsmaiTs enormous palace, now half in ruins, dom¬ 
inates much of the city. 

The French changed the character of Meknes not only by the addi¬ 
tion of the new quarter and their various military installations but also 
by attracting a huge rural population. In 1936 the population of 
Meknes was about 62,000—about 50,000 Muslims and the rest Euro¬ 
pean and Moroccan Jews. By 1960 there were nearly 161,000 in¬ 
habitants, 150,000 of whom were Muslims. Thus the average annual 
growth rate for the Muslim population of the city from 1936 to 1960 
was 4.35%, which contrasts with the Moroccan national average of 
2.5%. This difference is readily accounted for by rural immigration to 
the city (Service d’Urbanisme n.d.). 

Although the heterogeneous character of the population of Meknes 
may be attributed to this influx of rural peasants from both the 
Atlantic plains and the Berber plateaus and mountains, it should be 
remembered that the population had always been mixed. Not only 
was the city a sort of border town between the Arab and Berber 
worlds as well as a stop on the east-west caravan route; it was also a 
created city, to use the expression of the French urbanists, whose 



78 


THE HAM ADS HA 


inhabitants were attracted to it or brought in by force from many 
parts of Morocco. It was not Moulay Ismail alone who was responsible 
for this; later rulers on occasion brought in whole tribes, or tribal 
segments, to protect the city from outside threat or from internal dis¬ 
order. These groups settled in separate quarters of the city with which 
they are still identified, although over the years inhabitants have mar¬ 
ried out of their group and moved to other quarters. 

Like other major cities of Morocco, Meknes is stereotyped by both 
Moroccans and Europeans. The Moroccans consider it to be one of 
their most beautiful cities, endowed not only with a temperate climate, 
clear air, and adequate rainfall, but with work opportunities both 
within the city and in the nearby rural areas. Prices are said to be 
lower there than in some of the other Moroccan cities. Meknes is not 
recognized for its piety as is Marrakech, for its learning as is Fez, for 
its “big business” and corruption as is Tangier, or for its international 
flavor as is Casablanca. It is considered, instead, to be the most “Moroc¬ 
can” of the major cities. 

To the European, particularly the French, the beauty of the city 
was taken more in stride, and it was recognized as a provincial farming 
center and military base which had excellent economic possibilities and 
was, despite its provinciality, not an undesirable city in which to settle. 
This picture, however, was greatly tarnished by the so-called evene- 
ments which followed Independence. Then the inhabitants of Meknes 
rose up and rioted, and they stormed some of the European homes and 
factories and killed a few of the Europeans. These events, which seem 
to have been unique to Meknes, caused panic among many of the 
Europeans who left the city. Meknes, in fact, has had a rather violent 
history, from the time of Moulay Ismail to the present day. (It was 
the scene of one of the few violent incidents in Morocco during the 
six-day Arab-Israeli war in June of 1967.) In part, this “tone of vio¬ 
lence” can be attributed both to the continued presence of military 
installations in the city—Meknes was always and still is an “army 
town”—and to the city’s unusually heterogeneous population, which 
was subject to internecine conflict as well as to straightforward envy 
of the wealthier Jews and Europeans. 

More than any other Moroccan city, Meknes is known for its popu¬ 
lar brotherhoods, especially for the internationally famous Isawiyya, 
who follow Meknes’ patron saint Sidi M’hamed ben flsa and who have 
lodges all the way across North Africa to Mecca. The ‘Isawa are 
known for their trances, in which they attack anything that is 
black or anybody wearing black and, like the worshipers of Dionysius, 




THE LODGES OF MEKNES 


79 


tear apart (the Bacchic, sparagmos) and eat raw (the Bacchic omo¬ 
phagia) their sacrificial victims (Jeanmaire 1951:259 et seq.; Dodds 
1966:270-282). Although the ‘Isawiyya were the largest and best- 
known popular order in Meknes, many of the other Moroccan orders 
were and still are represented in the city. Not only are the Isawa and 
Hamadsha still active, but the Jilala, the Miliana, the Tuhama, the 
Gnawa, and even the Heddawa continue to meet there. Among the 
“higher orders,” the Tijaniyya and the Darqawiyya are still well rep¬ 
resented. None of these orders are as active as they were before the 
arrival of the French. Some of them, like the Miliana, have lodges in 
the madina which are now closed; the zawiyas of others, including the 
Hamadsha, are moribund. 

The Hamadsha Zawiyas 

The two Hamadsha zawiyas are located a few hundred yards apart 
in one of the poorest sections of Meknes, not too far from the smiths’ 
quarter. There are a number of butcher shops in the area, as well as 
shops where one can buy second-hand clothes, cord and rope, musical 
instruments, and odd bits of junk, or have an old jallaba mended or 
shoes repaired. It is not a specialized quarter, and the everyday needs 
of most of the local residents can be met within a radius of a few 
hundred yards. It is in easy walking distance of a little food market 
near Sidi ben ‘Isa’s mausoleum before which the Hamadsha of the 
shantytowns often perform on Fridays, along with adepts from other 
popular orders. 

Both zawiyas consist of enclosed courtyards in which grave plots 
have been sold to wealthy Arabs anxious to facilitate their entry into 
heaven by partaking of some of the baraka inherent in the place. This 
practice is not restricted to the Hamadsha. One finds graves not only 
in many of the lodges of the city, but also in the countryside in the 
vicinity of a saint’s tomb. Usually the founding saint is reported to 
have spent some time at the site of the zawiya or one of his descendants 
is said to be buried there. 2 The latter is true of the Hamadsha zawiyas 
of Meknes. At about the turn of the century, both orders were able to 
obtain their lodges by promising the revenues from the sale of grave 
plots to their previous owners. Until then they performed in the open 
or in private houses, much as the shantytown teams do today. Mem¬ 
bers of the Hamadsha brotherhoods themselves are usually too poor 

* Cf. pp. 25 et seq. 




80 


THE HAMADSHA 


to buy the expensive plots, and only one or two of them are buried 
there. The c Allaliyyin’s zawiya is slightly larger than the Dghu- 
ghiyyin’s, and has a room as well as covered porticos on two sides of 
the courtyard. The Dghughiyyin have only a portico on one side of 
their lodge. Both have fig trees in the center of the court which are 
dedicated to c A'isha Qandisha. As to the fig tree at Beni Rachid, bits 
of cloth have been tied to these trees as signs of various promises to 
the she-demon, most often a promise (*ar) to sacrifice a chicken to 
c Aisha if a child is bom. Many of the lodge members claim to have 
seen c A’isha in front of the trees. 

The organization of both lodges is much simpler than that of many 
of the other Moroccan brotherhoods. The Hamadsha zawiyas come 
under the direct charge of their particular mizwar, who not only re¬ 
ceives the money they collect for supposed distribution among the 
saints’ children but acts as an arbiter in cases of disputes among the 
lodge members or between the lodges of his brotherhood. He approves 
the choice of a local leader, a muqaddim, who is selected by the adepts 
themselves, and provides the muqaddim with both a letter of authority 
and a flag. In the case of the c Allaliyyin, the mizwar also gives the 
muqaddim a handwritten copy of the hizb (sacred writings) allegedly 
composed by al-Shadhili. The mizwars of both orders exercise con¬ 
siderably more authority over their Meknes lodges, which are close 
to the Zerhoun, than to zawiyas in more distant cities. This breakdown 
in central authority is characteristic of most Islamic brotherhoods and 
has severely limited the political potential of such great orders as the 
Tijaniyya or even the flsawiyya. 

Ideally, the muqaddim is unanimously chosen by the foqra, or 
adepts, who meet at the zawiya about a month or six weeks after the 
death or retirement of the previous muqaddim. The lodge members 
discuss the possible candidates, who may be present but usually arrive 
later, after a decision has been reached. “The muqaddim must be in¬ 
telligent, a good judge of people, and not overly powerful, and he 
must have the ability to speak to people at all levels,” an adept of one 
of the zawiyas explained. When the adepts have come to an agree¬ 
ment—there is no formal voting—they say a fatha, or prayer, over 
their new leader, and send word of his appointment to the mizwar. 
A few days later, the new muqaddim usually sponsors a ceremony to 
celebrate his election. 

In fact, the appointment of a new muqaddim usually involves con¬ 
siderable politicking, in the course of which various delegations of 
foqra visit the mizwar to press for their particular candidate. The 



THE LODGES OF MEKNES 


81 


mizwar then makes a decision based upon the popularity of the candi¬ 
date, the strength of his loyalty, his knowledge of the litanies, his 
leadership qualities, and, above all else, his honesty. He lets word of 
his preference leak down to the adepts, who then meet to “decide” 
upon their new leader. The preferred candidate, who is well aware of 
the mizwar’s choice, does not arrive until after the foqra have had a 
reasonable amount of time to make their decision, and he acts surprised 
when he learns of it. He declines at first but, upon the insistence of 
the lodge members, he finally permits them to say a fatha over him 
and declare him their new leader. The mizwar is then informed of the 
“decision” and sends the muqaddim a letter of congratulations and 
authority and the flag. The muqaddim is then expected to sponsor a 
ceremony in which litanies are chanted, the hadra performed, and a 
meal served. The ceremony serves to announce his appointment to 
important local personages like the muqaddim of the ‘Isawiyya and 
the foqra of the other Hamadsha brotherhood. It also helps the muqad¬ 
dim consolidate his position in the lodge, for his rivals and their sup¬ 
porters, having accepted his hospitality—shared his bread—are obliged 
to befriend and support him. 

In the spring of 1968, die muqaddim of the ‘Allaliyyin died. He had been 
leader of the zawiya for over twenty years and, well-established in his 
position, he had been able to cheat the children of Sidi c Ali out of much of 
the zawiya’s proceeds. Although some of the adepts respected the muqad¬ 
dim, many others disliked him. Several of these, like Sidi Mohammed 
Touijer, had begun to attend the meetings of the Dghughiyyin. For several 
weeks after the muqaddim’s death, there was much speculation among the 
adepts of both lodges as to who would become the new muqaddim. One 
group of ‘Allaliyyin supported the assistant ( khalifa ) to the former muqad¬ 
dim, a seller of amulets and leather scraps; a second a man named Ahmed, 
a butcher’s assistant; and a third, Touijer, a tanner. Touijer, a man in his 
thirties, was the youngest of the three. Each group of supporters sent a 
delegation to the mizwar at Beni Rachid to push their candidate and to 
disparage his rivals. The supporters of the khalifa argued that their can¬ 
didate was older and more familiar with the traditions of die zawiya than 

either Touijer or Ahmed and that Touijer was a Dghughi at heart and 
Ahmed was a drunk. Touijer’s supporters argued that their candidate knew 
the litanies best and that he was very honest. They maintained that the 
khalifa was as dishonest as the former muqaddim and that Ahmed was a 
drunk. Ahmed’s supporters used similar arguments. 

The mizwar told the different delegations that he would have to think 
the matter over and would let them know. He then went secretly to the 
muqaddim and khalifa of the Dghughiyyin and asked them their opinion. 




82 


THE HAMADSHA 


They were very pleased that he had asked them. The mizwar himself was 
anxious to improve relations with them—relations had deteriorated ever 
since the former muqaddim had been appointed leader of the ‘Allaliyyin. 
The two leaders of the Dghughiyyin urged the muqaddim to choose 
Touijer. The mizwar decided finally to appoint Touijer temporary muqad¬ 
dim (and Ahmed khalifa), and sent a letter to this effect to his lodge in 
Meknes. This served to avoid a “mock election” which, under the circum¬ 
stances, might have been embarrassing. It served also as a test for Touijer. 
His appointment would be validated, the mizwar said, immediately after the 
musem. In other words, Touijer would have to collect enough money to 
make his appointment worthwhile to the mizwar. Touijer, who was a hard 
worker, was familiar with the ways of not only the conservative adepts of 
the madina but also with those of the rural immigrants to the city, who had 
until then gravitated toward the Hamadsha of the bidonvilles. Thus he was 
able to collect enough money to please the mizwar, and was appointed 
permanent muqaddim. 

In August, about six weeks after his appointment, Touijer sponsored a 
ceremony to which he invited not only the £ Allaliyyin but also the Dghu¬ 
ghiyyin and the muqaddim of the Isawiyya. The latter, an educated man 
who acts unofficially as the leader of all the muqaddims of Meknes, was 
unable to come but sent his khalifa. Midway through the ceremony, the 
khalifa of the Isawiyya and the Dghughiyyin excused themselves and went 

into a separate room where, “much to their surprise,” the muqaddim served 

them couscous and then handed a letter to the £ Isawi to read. It was the 
muqaddim’s letter of appointment. After the letter was read, all of the 
Dghughiyyin said a fatha over him, and the new muqaddim gave them the 
first proceeds of his night, which was a substantial offering of over 60 
dirhams. They then returned to the ceremony. 

The muqaddim is charged with administration of the lodge; the col¬ 
lection of alms; the organization and direction of ceremonies; curing 
the sick during ceremonies and at other times; and settling minor dis¬ 
putes between the adepts. He is the liaison between the zawiya and 
the mizwar and between the zawiya and the municipal government. 
He is considered an authority on Hamadsha lore and leads the foqra 
in the recitation of the litanies. He is assisted by a khalifa, who may 
act also as his messenger. The khalifa is charged with keeping the 
supplies of the lodge and with letting all of the foqra know the 
schedule of ceremonies. The muqaddim and khalifa of the £ Allaliyyin 
both keep records of the alms and other proceeds from ceremonies 
that have been collected; the muqaddim safeguards the proceeds. 
Among the Dghughiyyin, a treasurer ( khzana ) keeps the money, and 
the muqaddim, the khalifa, and a registrar ( katib) each keep separate 



THE LODGES OF MEKNES 


83 


records as a check on the treasurer and on each other. These records 
are shown to the mizwar when the proceeds are presented to him. 
Each zawiya also has two female caretakers who keep the lodge clean, 
care for the guests’ slippers at the lodge, and take charge of female 
participants during the ceremonies. The muqaddim, the khalifa, the 
treasurer, the registrar, and the caretakers are the only “officers” of 
the zawiya, and they receive compensation in one form or another 
from the mizwar. 

The Foqra 

The foqra —today there are about 15 in each zawiya—are not or¬ 
ganized into any hierarchy. They do not go through any form of 
initiation or receive any secret instruction. 3 They are simply men who 
attend the meetings at the zawiya and assist regularly at the cere¬ 
monies. They may, in fact, belong to other orders as well, although 
none of the present-day adepts do. They usually dance the hadra and 
fall into trance, but dancing and trancing are by no means qualifica¬ 
tions for membership, nor is any detailed knowledge of the litanies. 
Many of them play the hourglass-shaped drum, the gwal. Most of the 
foqra will have invited their fellow lodge brothers to their homes 
for a sadaqa —a seance at which litanies are recited, the hadra per¬ 
formed, and a meal served—shortly after beginning to attend the lodge 
with any regularity. The sadaqa will usually have been accompanied 
by the sacrifice of a male goat or a ram. They all consider themselves 
to be foqra of the zawiya and followers of its founding saint, from 
whom they derive their good fortune and baraka which enables them 
to perform the hadra and experience the hal. 

The nature of an adept’s relationship to his saint is complicated. 
Many Moroccans feel tied to a particular saint who serves them as an 
almost personal intermediary with God. They speak of themselves 
as supported (msannad ) by the saint. In some instances the relation¬ 
ship may be formalized, but this is not necessary. A man’s hair may 
have been cut for the first time in the saint’s tomb. His parents may 
have taken him (or her) to the tomb when he was first born and 
“given” him to the saint. If there are descendants of the saint, the 
child may be presented to their leader for blessing. The leader says 

3 Although there is no initiation ceremony in either the lodges or teams of 
Meknes, there may be initiatory practices elsewhere. Schoen (1937:12, fn. 1) 
claims that the muqaddim’s spitting on the head of a new Hamdushi serves as an 
initiation. 



84 


THE HAMADSHA 


the fatha over the child—he may spit or blow on it—in exchange for 
money or a gift from the parents. The presentation of the child to the 
saint or to his descendants usually occurs only if the child was bom 
after the parents had made a pilgrimage to the saint to ask for one. If 
a man was never formally presented to a saint, he may still in later life 
obtain the saint’s support by making a pilgrimage to his tomb, asking 
the saint to be his support, and leaving an offering. Often, the inspira¬ 
tion for this is a dream which occurs during a crisis in the man’s life. 

H. was apparently worried because he had publicly supported the King of 
Morocco during his exile and had been warned that he would be thrown 
into prison if he continued. He had a series of dreams about Moulay Idriss, 
the last of which involved a man who had appeared in earlier dreams. The 
man gave him a white candle. H. asked the man what he was going to do 
with the candle. The man said to bring it to him. The next day H. left for 
Moulay Idriss and prayed. He did not ask to become msannad, but when 
he left the tomb he found IVz dirhams in front of it and has considered him¬ 
self msannad by Moulay Idriss ever since. He stresses the fact that he did 
not choose Moulay Idriss but Moulay Idriss called him. The man in the 
dream was Moulay Idriss. 

Although many Moroccans arc not msannad by any particular saint, 
and although the formalization of such a relationship is neither re¬ 
quired nor fixed, once a man is supported by a saint he cannot be 
msannad by another. This does not mean that he cannot visit other 
saints to ask their help. He is always privileged to do this. 

Neither the descendants of Sidi c Ali or Sidi Ahmed nor the foqra 
are necessarily supported by either Sidi c Ali or Sidi Ahmed. Many of 
the saints’ descendants have had their first haircuts in the tomb of 
one of the lesser Zerhoun saints. The foqra consider themselves to be 
followers ( taba'in ) or workers ( kheddemta) of one of the Hamadsha 
saints, but they may be msannad to other saints or to no particular 
saint at all. Those who arc msannad to one of the Hamadsha saints 
usually come from a family in which there were already either 
Hamadsha or members msannad by the saint; they were most often 
presented to the mizwar for his blessing. Hamadsha who wanted their 
sons to become head-slashers occasionally brought them to the tomb 
of one of the saints and pressed their heads down on a miniature of 
the axe with which the Hamadsha traditionally slashed their heads. 

The distinction between being supported by a saint and being a 
follower of one is not altogether clear. Ideally, one can obtain every¬ 
thing one wants, Allah willing, through either one’s supporting saint 



THE LODGES OF MEKNES 


85 


or the saint one follows (or, for that matter, from any saint). The 
main difference appears to be one of focus and exchange. The fol¬ 
lower works for the saint he follows; that is, he participates in (or at 
least attends) the Hamadsha ceremonies, the proceeds of which are 
given to the saint. He has unwittingly obtained the baraka of the 
saint, much in the same way that the Sufi mystic obtains the hal; it is 
descended from God (or the saint). It cannot be obtained by work; 
but once having obtained it, the recipient is obliged to work for the 
saint. He gives the proceeds from his work (the ceremonies) to the 
saint, or his descendants, in exchange for the baraka he has obtained 
from him. This baraka not only enables the follower to perform the 
hadra and experience the hal (in the Hamadsha sense of the word) 
but, so long as he works for the saint, it keeps him in good fortune 
and health as well. Strictly speaking, one does not work for one’s 
supporting saint unless one is also a follower. There are many people 
msannad by cither Sidi c Ali or Sidi Ahmed who never perform the 
hadra or attend a Hamadsha ceremony. The person who is supported 
by a saint finds his saint to be a moral and psychological support in all 
the domains of his life. Although the saint theoretically is only an 
intermediary to God, for the average illiterate Moroccan he is the 
source itself of comfort and solace, help and support. 

The position of the adepts in urban society is difficult to determine. 
Although LeToumeau has suggested that there is a correlation be¬ 
tween membership in the Hamadsha brotherhoods and in certain pro¬ 
fessions (1949:366), I was able to find no evidence of this in Meknes. 4 
The adepts do belong to the lowest economic strata of the madina— 
they earn less than five dirhams a day—and are engaged in occupations 
in which they not only use but often dirty their hands; they are either 
totally or nearly illiterate; but they are neither recruited from nor do 
they dominate any one occupational group. They themselves rec¬ 
ognize no connection between being a Hamdushi and having a cer¬ 
tain occupation. The diversity of their professional ties is illustrated 
in Table 1. And just as there appears to be no relationship between 
occupation and membership in the brotherhoods, so there seems to be 
no relationship between membership and place of residence or work. 
The lodge cannot be conceived of as either a “professional” or a neigh¬ 
borhood organization. 

The foqra, like the Hamadsha generally, have been stereotyped by 
many wealthier, more literate Moroccans as being black, and many of 

4 LeCocur (1968:98) claims that in Azzemour the blacksmiths, drivers, porters, 
and fishermen were affiliated with the hamadsha. 




86 


THE HAMADSHA 


TABLE 1 


Profession 

i Allaliyyin 

Dgbugbiyyin 

Total 

Baker’s assistant 


1* 

1 

Barber 

1 

1 

2 

Butcher 

1 

1 

2 

Carpenter 


1 

1 

Cloth merchant 

1 

1 

2 

Cook 

1 


1 

Gardener 

3 

1 

4 

Grain merchant 


1 

1 

Greengrocer 

1 

1 

2 

Jallaba merchant 


1 

1 

Kefta salesman 

1 


1 

Lime dealer 


1 

1 

Pallbearer 


1 

1 

Porter 

1 


1 

Shoemaker 

2 


2 

Tailor 


2 

2 

Tanner 

1 


1 

Weaver 

2 

1 

3 


15 

15 

30 


* The baker's assistant joined the army and now lives on a pension without 
working. 


them are often considered to be black. They are, nevertheless, differen¬ 
tiated from the members of the so-called black confraternity, the 
Gnawa, who claim to be descendants or followers of Sidina Bilal, the 
Prophet’s slave, and who until a few decades ago spoke African lan¬ 
guages. Since Moroccan notions of race arc social typifications and 
not biological ones, such statements as “The Hamadsha are black” 
must be understood not from a racial, or even an ethnic, point of view, 
but as a means of ascribing relative social status. In Morocco, a per¬ 
son’s racial identity varies according to his own social status and to 
the status of the person classifying him. 

It is possible—and there is some evidence for this—that historically 
many of the Hamadsha were recruited from among the black slaves 
who were brought north from sub-Saharan Africa, and that their 
stereotype as black also reflects this. The tribal affiliation of the two 
orders suggests this. (See Table 2.) The Filala, the Bukhara, and the 
Drawa, who make up 40% of the membership of both orders, may 






THE LODGES OF MEKNES 


87 


TABLE 2 


Tribt 

l Allaliyyin 

Dghughiyyin 

Total 

Berber 

2 

2 

4 

Bukhara 

3 


3 

Drawa 

2 


2 

Fez* 


1 

1 

Filala 

1 

6 

7 

Gharbawa 


3 

3 

Jbliya 

3 


3 

Meknes* 

1 

1 

2 

Ouazzana 

1 


1 

Taza* 

1 


1 

Zrahna (Zerhoun) 

1 

2 

3 


15 

15 

30 


* No memory of tribal affiliation. 


have come originally from sub-Saharan Africa. The Bukhara were 
brought to Meknes by Moulay Ismail as military slaves. 

The fact that only two of the foqra consider their families to be 
from Meknes, and that these are unable to recall their tribal origin, is 
indicative of the recent arrival of the majority of foqra to the city. 
(Genealogies tend to be shallow; three ascending generations appear 
to be the limit.) But unlike the Hamadsha of the shantytowns, who 
are primarily first-generation arrivals the foqra of the madina are 
second or third generation. 5 

Since more than half of the 'Allaliyyin and 80% of the Dghughiyyin 
are over 45, the families of the adepts appear to have been attracted 
to the city not by work opportunities created by the arrival of the 
French but by those created by pre-Protectorate city life in general. 
Many of the foqra explain that their parents or grandparents first came 
to Meknes because they were called in by the government to fill 
specific positions for which there was a shortage of men. 

Membership in the Hamadsha brotherhoods is not hereditary. Since 
both orders were apparently loosely organized before they obtained 

“Ten of the 15 'Allaliyyin were bom in Meknes, as were 9 of their fathers; 
9 of the 15 Dghughiyyin were bom there, 4 of their fathers, and only one of 
their paternal grandfathers. Information concerning the paternal grandfathers of 
the 'Allaliyyin is incomplete. Few, if any of them, appear to have been bom in 
Meknes. 





88 


THE HAMADSHA 


their zawiyas in the beginning of this century, and since many of the 
parents of the present-day adepts were not bom in Meknes, it is im¬ 
possible to determine whether or not zawiya membership followed 
family lines. There does appear to be a tendency for the adepts of both 
lodges to come from families in which at least one parent performed 
the hadra. This is true for 8 'Allaliyyin (7 fathers and 2 mothers, with 
an overlap of one) and 11 Dghughiyyin (11 fathers and 5 mothers, 
with an overlap of 5). Although my data for the second ascending 
generation are incomplete, they suggest that a significant number of 
paternal grandparents also performed the hadra. It should be pointed 
out, however, that there are 8 families among the adepts of both zawiyas 
in which the present-day adept is the only Hamdushi. 

Four of the wives of the 9 married ‘Allaliyyin and 3 of the wives 
of the 13 married Dghughiyyin participate in the hadra. A smaller 
number of children appear to be following in their fathers’ footsteps.® 
This is usually explained away as the result of education. The foqra 
themselves recognize that the Hamadsha come from among the illiter¬ 
ate, and it is not unusual to hear them say that so-and-so’s children 
do not perform the hadra because they have been to school. 

Apart from the regular adepts, who often play drums at a ceremony, 
there are professional musicians— ghiyyata } or oboe players—attached 
to the orders. The gbit a is the most important instrument of the 
hadra, and playing it requires considerable skill and endurance. The 
exact status of the ghiyyata in the Hamadsha brotherhoods has been a 
source of confusion to other investigators. Although the ghiyyata have 
been affiliated with the zawiya for years, they are treated as distinct 
from the foqra, both because they receive remuneration for each per¬ 
formance and because they ply their trade elsewhere as well. (The 
ghita, usually a smaller version of the ghita gharbawiyya used by the 
Hamadsha, is played at marriages and circumcisions and at the cere¬ 
monies of other brotherhoods like the Isawiyya and the Jilaliyya.) 
Moreover, their experience of the hal is different from that of other 
adepts, who may fall into a more frenetic trance called jidba. The 
ghiyyata, who must exercise great self-control, cannot enter such a 
state without disrupting the hadra entirely. 

•At least one child of 3 £ Allaliyyin—8 have children—and one child of 4 
Dghughiyyin—8 have children—participate in the hadra. It should be noted that, 
considering the age of the adepts, an above-average number of £ Allaliyyin (6 out 
of 15) are single and an above-average number of the married Dghughiyyin (5 
out of 8) are childless. 



THE LODGES OF MEKNES 


89 


There are two ghiyyata in each brotherhood; they play the guitar 
(ganbri) and recorder ( nira ) as well. Each of the four ghiyyata comes 
from a family in which members performed the hadra; one of them 
is a descendant of Sidi Ahmed Dghughi. Ghita playing is the only 
source of income for three of them; the fourth also sells rugs. The 
rug merchant comes from an old Meknes family. The other three come 
from the Jebel Zerhoun. 

Each lodge also has a professional ganbri player. This role is not so 
well differentiated as that of the ghiyyata, and it may be unique to 
the Meknes orders. Usually, the ghiyyata also play the ganbri. The 
two ganbri players come from the Gharb and earn their living by 
playing and begging. One of them attends the ceremonies of both 
orders regularly. They, too, are paid for their performances and are 
not considered foqra. 

The Muhibbin 

The last and largest group associated with the lodges is the muhib- 
bin, or devotees, of the order. The word “muhibbin” is derived from 
the Arabic for love (hub), and refers to all the men, women, and 
children who are in some manner attracted to the hadra of the 
Hamadsha and who attend the performances—usually of one or the 
other of the two brotherhoods—whenever they can. They do not 
necessarily meet at other times or under other circumstances. They 
are, as it were, the tertiaries of the order. The hadra (and, to a lesser 
extent, other ceremonies of the lodges) is the focal point of the muhib¬ 
bin; it is the one activity in which they all engage. The devotees form 
a group, as it is classically defined by sociologists, insofar as there is 
social interaction between them, but they differ from the “group” in¬ 
sofar as there is no clear-cut boundary between member and non¬ 
member. The fluid boundaries of the muhibbin arc typical not just of 
the Hamadsha but of other religious confraternities of the Islamic 
world. 

There are definite differences among muhibbin, which are rec¬ 
ognized both by the foqra and by the devotees themselves. Although 
it is possible to speak of a particularly active devotee as a muhibb 
kabir, a great devotee, there are no verbally differentiated categories 
of muhibbin. Differences arc expressed by tone of voice or by para¬ 
phrasis. There appear to be two distinct factors which determine what 
kind of muhibb an individual is: the intensity of his involvement with 




90 


THE HAMADSHA 


the hadra, and the frequency of his attendance at the hadra. The latter 
can easily be measured. Some muhibbin attend the hadra from three to 
four times a week; others, once or twice a year. Intensity of involve¬ 
ment is more difficult to determine. Some devotees perform the hadra, 
enter into a frenetic trance, and mutilate themselves; others simply 
like to attend the ceremonies but do not participate. Figure 2 presents 
a somewhat arbitrary summary of the degree of involvement of the 



FIGURE 2. 



































































































THE LODGES OF MEKNES 


91 


muhibbin. There are 9 measures of intensity and 4 measures of fre¬ 
quency, providing 36 possible categories of muhibbin. (An understand¬ 
ing of all of the significant measures will be possible only after an 
analysis of the hadra and patterns of trance has been made.) All of 
the factors listed are recognized by the Hamadsha themselves, although 
the compulsive quality of some of the activities is recognized by only 
the more perceptive. 

There are no formal ties between the muhibbin and the zawiya. 
The foqra know many of them and invite them to their ceremonies, 
but all of the devotees do not attend any one performance. They 
form clusters of individuals who attend one another’s ceremonies. 
Only the most active devotees are present at all of a zawiya’s perform¬ 
ances. The composition of these clusters is informal; they consist of 
both patrilateral and matrilateral relatives, friends, neighbors, work 
associates, and other devotees respected for their piety, their devotion 
to the founding saints of the orders, or the intensity of their hadra 
(as measured by the factors in Figure 2). The clusters overlap and 
may best be conceived as an aggregate of individual networks in 
which there is a high incidence of recurrent membership (cf. Barnes 
1968:118). A network—a primary order star in Barnes’ terminology 
—is limited here to all of the devotees whom an individual will in¬ 
vite to ceremonies he sponsors. The network has, of course, no external 
boundaries and no clear-cut internal divisions. 

Boualem and His Network 

Boualem is a typical madina muhibb of the Hamadsha, although at 30 he 
is perhaps a litde younger than average. He was bom in the madina of 
Meknes. His paternal grandfather, a member of the Chaouia tribe from the 
countryside around Casablanca, came to Meknes as a young man. Boualem 
still considers himself a member of the Chaouia, although he has never 
visited his tribal “homeland.” He has been told that not only is there the 
belief in ‘ATsha Qandisha and other jnun there, but also a spot sacred to 
Sidi e Ali ben Hamdush. It is said that Sidi e Ali spent some time there on his 
way to Bu'abid Sharqi in Marrakech. This spot, which is marked by a 
qubba and a sacred fig tree, is visited by people suffering from headaches, 
paralysis, and other “problems” caused by the jnun. Women anxious to 
have children also visit it. Not only are there Hamadsha in the area but also 
Tsawa, Jilala, and Darqawa. Most of the members of Boualem’s immediate 
family are Hamadsha, as the partial genealogy in Figure 3 indicates. 

His parents often invited the Hamadsha to their house; his father, a peasant, 
was particularly prone to attacks of the jnun which took the form of paraly¬ 
sis, and could only be cured by the hadra. 



92 


THE HAMADSHA 


A 


O 


£ I.I 




A or# IJJADRA PERFORMER A HEAD-SLASHER 

FIGURE 3. 

Boualem went to Koranic school for only seven months, and he never at¬ 
tended government school. He is capable of signing his name, but can 
neither read nor write anything else. He works for a Frenchman in a wine 
cellar and claims that among the Frenchman’s workers there are many 
muhibbin. He never worked as an apprentice for a master artisan and, like 
other madina inhabitants, he works the harvest in the vicinity of Meknes 
whenever possible. Boualem, who has always lived in Meknes, now rents a 
house in the ??iellah, or Jewish quarter, of the city, to which many Arabs 
moved after the Jewish exodus in the late fifties. Rents are cheaper there. 
Boualem is married to a Berber who was bom in Meknes. (Arab-Berber 
marriages are still the exception, even in the city.) He has three children, 
the oldest of whom is 7 and goes to school. They are too “small” to per¬ 
form the hadra. 

Boualem has made several pilgrimages to Sidi c Ali and has experienced the 
hal and performed the hadra. He was 15 years old when he first danced. 
He had a fever at the time, his bones ached, and he felt as though “water 
were boiling in his head.” He slashed his head. It had begun to sting and 
felt as though there were a “worm” in it. Since then, he tries to dance at 
least four times a month. (His rih is from the Zcrhoun; his color is black; 
and c Aisha Qandisha is his jinniyya.) 1 He seldom slashes his head now. He 
attends the performances of the ‘Allaliyyin, but as they are not active 
enough today to satisfy his needs, he also attends performances given by 
a team from a little town just outside Meknes. If he does not perform the 
hadra often enough, his body begins to tremble, his hands become heavy, 
his knees shake, and he is vulnerable to attacks by the jnun. He, or his 
parents, invites the Hamadsha at least once a year. 

’These terms will be explained later. 






THE LODGES OF MEKNES 


93 


Boualem has a number of acquaintances who are muhibbin. He invites them 
to his ceremonies, and they invite him to theirs. Fourteen devotees—those 
with whom Boualem immediately associates both socially and psychologi¬ 
cally—were interviewed. 8 Two of them were his brothers. They all con¬ 
sider themselves muhibbin; with one exception, they all fall into the first 
twelve categories of devotees. This reflects the fact that the term “muhibbin” 
refers primarily to the most active—the most intensely involved—Ha- 
madsha. No women are included in the group. Although there are undoubt¬ 
edly certain women who usually attend Boualem’s ceremonies, they are, 
owing to the segregation of the sexes, part of Boualem’s wife’s or mother’s 
network. Boualem himself cannot invite them. Invitations must pass either 
through his wife or their husbands. 

Figure 4 summarizes the significant facts concerning the members of 
Boualem’s network. The members all live in the madina, but in no single 
quarter. With two exceptions they are all about Boualem’s age and are 
probably somewhat younger than the average muhibb. Their age also ac¬ 
counts for the high percentage who are single. The average muhibb is 
married. The fact that only 5 of them were bom in Meknes is also excep¬ 
tional. Of the 12 devotees who are not related to Boualem, only one is of 
his tribe; the majority of them are not related either by blood or marriage. 
They tend to work in the same area of Meknes; tanners are well repre¬ 
sented. It appears that occupational ties are here not altogether insignificant 
—even though, in this case at least, they are established not through the 
nodal figure of the network but through his brothers, both of whom are 
themselves tanners. It is possible then to state that the muhibbin in Boua¬ 
lem’s network are about his age and occupy social positions which are sim¬ 
ilar to his both in terms of the type of work and degree of integration into 
the urban milieu. 

With one exception, all of the devotees perform the hadra; 6 of them slash 
their heads. They all appear to need to dance at least three times a month, 

'These devotees may be conceived of as forming a network (a primary order 
star) with Boualem as the center (Barnes 1968:113). This network must not be 
reified or considered as an indigenous conceptualization of a social reality. It is 
an analytic device, employed by the investigator, and must be taken as such. 
Boualem is in no sense a leader of the network; he has simply been taken as a 
reference point for a particular investigation. The 14 persons interviewed do not 
comprise Boualem’s complete network; a network is in the final analysis boundless. 
Rather they were chosen because they were the devotees he invites to his nights. 
(It was impossible to ascertain how frequently he invited them.) He gave the 
same names when asked who invites him to their nights; however, as a second 

? uestion, this answer was, in a sense, prefigured. Boualem probably suspected that 
would want to interview them and may, therefore, have chosen those with 
whom he was especially friendly. Although this method of inquiry is not alto¬ 
gether satisfactory, it does, in my opinion, give an adequate picture of the muhibbin 
and their social relations. It is noteworthy that Boualem did not list his father, 
with whom he often sponsors ceremonies; he did name his brothers. 



94 


THE HAMADSHA 


m 

m 

r- 

D 

X 


u 

o 

< 


8 

< 



o h o 
HZ joz 

£ 2 u. 2 
u. c . q: 

*8*8 
u a: ic a: 

Ul (il tj iiJ 

2°- io°- 


x- 

£ 


EGO 

MEKNES 

CHAOUIA 

25-30 

WORKS IN 

WINE CELLAR 

+ 

+ 

15 

+ 

+ 

3 

IJl 


CHAOUIA 

30-35 

TANNER 

□ 

B 

Qj| 


B 

B 

BR-2 

MEKNES 

CHAOUIA 

20-25 


H 

B 

B 

B 

B 


m 

ZERHOUN 

C ALAWA 

25-30 

TANNER 

S 

B 

B 

B 

B 

m 

5 

TARA 

BRANES 

50 + 

PEASANT 

□ 

B 

B 

B 

B 

B 

n 

MEKNES 

TSOUL 



□ 

B 

B 

B 

B 

B 

m 

MEKNES 

REHAMNA 

25-30 

BARBER 

D 

B 

B 

B 

B 

B 

n 

MEKNES 

FILALA 

15-20 

TANNER 

- 

- 

B 


m 

B 


ZERHOUN 

SKHIRA 

60-1- 

BUTCHER 

D 

B 

B 

— 

B 

B 


ZERHOUN 

HAMMADA 

25-30 


B 

B 

m 

n 

B 

B 


MEKNES 

CHAOUIA 

25-30 

TANNER AND 
COPPERSMITH 

B 

B 

B 

B 

B 

B 

D 

ZERHOUN 

OULED YOUSSEF 

35-40 

COPPERSMITH 

- 

+ 

,8 _ 

B 

B 

B 

D 

MEKNES 

ZRAHNA 

15-20 

BARBER 

B 

B 

B 

m 

B 

B 

D 

MEKNES AREA 

o 

25-30 

PEASANT 

B 

B 

B 

B 

B 

B 

D 

MEKNES AREA 

■> 

40-45 

porter 

B 

B 

m 

B 

B 

B 


FIGURE 4. 


and they all express the same symptoms of heaviness, depression, and 
lethargy if they do not dance. The hadra appears to invigorate them. Eleven 
of the 14 started to dance between the ages of 10 and 20; 17 is the average 
age. Six of them were, like Boualem, sick before they fell into trance for 
the first time. This sickness was said in all cases to be caused by the jnun. 
Thirteen report that other members of their family perform the hadra; 
10 had at least one parent who danced. All of them have made pilgrimages 
to the founding saints of the Zerhoun. Most of them perform the hadra 
publicly as well as privately; they or their parents have all invited the 
Hamadsha to perform. It is significant that they do not all attend the 














































THE LODGES OF MEKNES 


95 


g 2 <n 

< a to 

9 « i- 

o: 1 2 

•** u. 2 

< ° i 

£ o z > 

5 z 2 oc 

n iu s o 

£ § £ 8 


«<5 8 | 

£ I X. 

uf 


|X- fe 

5 « 111 
U2 a. 

>j£ 

io 

2UI i- < 
«t 0- < I- 


L a: w 


x a u o „ 

i- < it. < u. 

I 1 * 1 ’ * 


B 

C A 

— 

A 

4 

Z 

+ 

+ 

+ 

4- 

z 

B 

C A 

B 

C A 

+ 

A 

3 

3 

+ 

+ 

+ 

S 

na 

na 

na 

G 

C A 

4- 

A 

3 

7 

+ 

+ 

+ 

S 

na 

na 

na 

B 

C A 

— 

0 

3 

2 

+ 

— 

+ 

s 

na 

na 

na 

R 

*H 

•f 

A 

8 

6 

— 

— 

— 

-1- 

M 

Y 

M 

R 

C A 

-1- 

A 

2 

7 

+ 



s 

na 

na 

na 

B 

CA 

+ 

0 

8 

7 

+ 

— 

+ 

s 

na 

na 

na 

no 

i*a 

no 

— 

na 

32 

+ 

— 

— 

s 

na 

na 

na 

— 

— 

+ 

6 

3 

4 

+ 


— 

T 

•> 

? 

? 

B 

C A 

+ 

0 

3 

12 

-1- 

T 

4 

s 

na 

na 

na 

B 

C A 

+ 

0 

IO+ 

4 

+ 

+ 

4- 

s 

na 

na 

na 

B 

c A 

+ 

A 

3 

4 

+ 

— 

— 

— 

na 

na 

na 

B 

c A 

+ 

A 

4 

4 

+ 

T 

+ 

s 

na 

na 

na 

B 

ca 

+ 


S 

7 

+ 

+ 

~ 


na 

na 

na 

B 

C A 

— 

— 

6 

7 

+ 

+ 

— 

— 

na 

na 

na 


A c allaliyya 
c a c aTsha 

B BLACK 
D DGHUGHIYYA 
G GREEN 
GH GHARBAWI 
M LALLA MIRA 

no NOT APPLICABLE 

S SINGLE 
SH SIOI HAMMU 
Wl WIFE 
Y YELLOW 
Z ZERHUNt 


FIGURE 4. 

ceremonies of a single zawiya. Many of them, like Boualem, go to the 
ceremonies given by other teams. 

Perhaps the feature that the devotees have most in common is the Zerhoun 
rih 9 and devotion to £ AIsha Qandisha. It is curious that despite considerable 
tribal representation all of the members of Boualem’s network, with the 

•The rihy which is examined later in this study, is a tune, or musical phrase, 
that will send a particular adept or devotee into trance. Each person is believed 
to have such a tune, which is associated with a particular jinn and a particular 
color. The rihs fall into several groups, named alter tribal regions; but response 
to them is not restricted to tribal members. 




96 


THE HAMADSHA 


exception of one of his brothers, have a trance-inducing tune from the 
same musical group. Whether this is accidental or caused by some social 
or psychological predisposition must be left an open but by no means in¬ 
significant question. 

A General Picture 

Boualem and the other members of his network are not unusual 
muhibbin. They exhibit many of the social and psychological char¬ 
acteristics of the average muhibb. Such a devotee, at least one who falls 
into the first 12 categories of Figure 2, refers to himself as a 
muhibb and considers himself a follower of either Sidi c Ali or Sidi 
Ahmed. He does not speak of himself as working for the saint, as the 
adepts do. He is usually devoted to e A‘isha Qandisha as well. 10 He 
has made at least one pilgrimage to Beni Rachid or Beni Ouarad, and 
he makes every effort possible to attend the annual pilgrimage. He 
docs not play an instrument. Occasionally he attends the Friday after¬ 
noon ceremonies (see below) at one of the zawiyas, and the private 
ceremonies of either zawiya whenever he is invited. If he must dance 
frequently—three times a month or more—he usually attends ceremo¬ 
nies performed by teams not affiliated with the zawiyas. He may even 
invite them himself. This results from the moribund condition of the 
zawiyas and the more spectacular performances of the teams. 

He is usually bom in Meknes, but can still state his tribal origin. If 
he is older than the members of Boualcm’s network, his family has 
come to the city at least a generation before him. He will have a trade 
in which he works with his hands and often dirties them. He earns about 
five dirhams a day and belongs to no particular guild or occupational 
group. If he falls within the age range of Boualem’s network, he tends 
to be less integrated into the urban milieu. He is usually completely or 
quasi-illiterate. He goes to the mosque on Friday but, like many Mo¬ 
roccans, seldom performs his daily prayers. There are usually other 
members of his family who have performed the hadra, often one of his 
parents. His family comes from an area in which there was a belief in 
e ATsha Qandisha and in which there were Hamadsha, but from no one 
tribal group. Often—perhaps more often than in Boualem’s network— 
he traces his devotion to the Hamadsha to an illness, usually conceived 
in terms of being struck or possessed by ‘A'isha Qandisha or other jnun, 
which was cured either by the Hamadsha themselves or by a pilgrim¬ 
age to the tombs of the founding saints. This illness often appears to 

10 The exact nature of this relationship will be examined in Part Three. 



THE LODGES OF MEKNES 


97 


the Western observer as hysterical in nature. If his trance dance has 
not been “a good one”—that is, a satisfying experience—or if he has 
not danced in a long time, he suffers a mild depression, a general ma¬ 
laise, and parasthetic pains which he describes as “pinching bones.” 
Sometimes he experiences a recurrence of the symptoms which first 
caused him or his family to invite the Hamadsha to treat him. 

The members of his network are all men and tend to be of his age. 
There is considerable variation in his relationship to them, although 
they often work in the same quarter as he and may often be members 
of his profession or a profession which is centered near his own. They 
may be members of his tribal group, and usually at least a few of them 
are related to him. It is impossible to state whether they, like the mem¬ 
bers of Boualem’s network, share the same rih, although they have in 
common their devotion to c Aisha Qandisha. 


The Ceremonies of the Zawiya 

There are three principal ceremonies which are given by the adepts 
of the zawiyas and which provide the zawiyas with the funds necessary 
for their maintenance and for their gifts to the children of their respec¬ 
tive saints. One of them—the lila , or night—often serves as a curing 
ceremony, and is treated in detail in the third part of this work. 

Of the other two ceremonies, the first is the Friday afternoon meet¬ 
ing. It is very informal and subject to considerable variation. The 
doors of the zawiyas, which are usually kept locked, are opened every 
Friday afternoon—the Muslim Sabbath—to both the foqra and the 
general public. The regular members of the lodge gather there to drink 
tea, gossip, and chant a few of their litanies—whichever litanies they 
know best and like the most. 11 The adepts do not perform the hadra 
there because, they claim, the entire floors of both lodges are “covered 
with tombstones.” It appears that the foqra used to perform the hadra 
on Fridays but that in recent years they have given up the practice, not 
only because of the tombstones but because of their age. Friday hadras 
are now performed by the bidonville teams in front of Sheikh al- 
Kamal. 12 


“The manner of recitation and the content of the litanies is discussed in the 
section on the night. Litanies recited at the zawiyas arc the same as those recited 
at the nights, but they are repeated in a calmer manner. They are not designed 
to send the adepts into an ecstatic transport. 

“Before the adepts had lodges, they used to meet in the cemetery near Sheikh 
al-Kamal on Fridays to perform. At tnis time there seems to have been no concern 



98 


THE HAMADSHA 


There are usually some male visitors who stop to leave a few francs, 
a candle or so, or even occasionally a chicken, but the majority of Fri¬ 
day visitors are women who come to glean some of the baraka of the 
place, to have a prayer said for them by the muqaddim or another 
brother, and to listen to the litanies. 13 Apart from acting out general 
devotion to the saints and ‘Ai'sha Qandisha, who is often said to oblige 
an individual to leave something at the zawiya each week, the women 
also usually come to ask for something—a child, relief from stomach 
pains or “pinching bones,” the preservation of a marriage, or simply 
good luck. They too leave candles, bread, sugar, or couscous. Often, 
especially at the lodge of the Dghughiyyin, they will tie bits of cloth 
to the branches of the fig tree as a sign of a vow ( c ar ) to do something 
for the saint or c Aisha Qandisha—usually sacrifice a chicken—if what 
they asked for comes true. Such visits may be likened to the pilgrimage 
to the saint’s tomb. 

Although the Friday afternoon ceremonies are a constant source of 
income for the lodges, the second ceremony, the name-day ceremony, 
a regular feature of the Moroccan life-cycle which takes place seven 
days after a child’s birth, is said to be the zawiya’s chief source of in¬ 
come. The foqra of the Hamadsha brotherhoods—like those of other 
brotherhoods—are often invited to the name-day festivities of wealthy 
muhibbin and other wealthy madina inhabitants who wish their chil¬ 
dren to be msannad by either Sidi e Ali or Sidi Ahmed or, at least, to 
glean some of the baraka of the saints. The adepts repeat the fatha over 
the newborn child and often chant litanies. This is said to give the 
child some of the saint’s baraka. The foqra usually attend the sacrifice 
of a male goat or ram and partake of the communal meal at the end of 
the festivities. 

If a couple has been able to have children only after a visit to either 
Sidi ‘Ali or Sidi Ahmed, the foqra who work for the particular saint 
held responsible for the child’s birth may be asked to name him. This 
is considered a great honor by the adepts and a guarantee of continued 
good fortune for both parents and child. If an infant is bom to an 
adept, his father will usually invite the foqra of his zawiya to perform 
the hadra as well as to recite prayers and litanies. The members of his 


about dancing over the graves of the dead; the older adepts, however, maintain 
that the government stopped their performances because they took place in the 
cemetery. 

“It must be remembered that women in the Islamic world do not attend the 
services at a mosque. Worship of saints and associated cults provides women with 
some “communal religious activity.” 



THE LODGES OF MEKNES 


99 


lodge will usually send him some sugar in the morning before the 
ceremony, to help him pay for it. He does not have to pay them, but 
the lodge receives all of the alms collected during the ceremony. Often, 
to feed all the guests, the adept will have to borrow money from the 
zawiya and return it when he can; he usually adds a few extra dirhams 
when he repays this loan. If a muhibb or someone else not related to 
the zawiya invites the lodge members, they must pay the lodge between 
20 and 50 dirhams for the ceremony. 

Distribution of Income 

The money and gifts collected at the ceremonies are theoretically 
for the children of the saint as a genealogical extension of the saint 
himself. They are held by either the muqaddim of the zawiya, as in the 
case of the c Allaliyyin, or the treasurer, as in the case of the Dghu- 
ghiyyin, until the annual pilgrimage, when they are given to the miz- 
war for distribution to the children of the saint. In the course of the 
year, the money may be lent to foqra, or to friends of the foqra by way 
of a particular adept, to tide them through periods of financial strain 
such as a birth, a circumcision, a marriage, an illness, or a death in the 
family. Although no interest is charged for such a loan, it is expected 
that the borrower will return somewhat more than he borrowed. If 
the adept is particularly poor, he is given the money outright. In this 
way, the zawiya functions as a society for mutual protection. 

In fact all of the alms and gifts received by the lodge do not go to 
the mizwar. Aside from the money given outright to the poorest adepts 
when they are in financial crisis, many of the gifts are perishable and 
so are distributed immediately to the foqra and the other poor. Each 
zawiya has its own rules, but the differences between them are of little 
importance. 

The alms and gifts received by the Dghughiyyin are distributed 
in the following manner. The money given the adepts at the Friday 
meetings is left with the treasurer for safe-keeping; the amount is regis¬ 
tered with the muqaddim, the khalifa, and the registrar. Usually, before 
counting it, a few francs are distributed to the poorest foqra. (In fact, 
all of the foqra receive a little.) The muqaddim, the khalifa, the treas¬ 
urer, and the caretakers do not receive any. Couscous is eaten imme¬ 
diately by the adepts; bread is distributed among the poorest foqra 
and among beggars who come to the lodge on Fridays for alms. 
Chickens are given to the families of two of Sidi Ahmed’s descendants 
who live in Meknes. Sugar is saved by the treasurer for ceremonies 



100 


THE HAMA DS HA 


like the name-day ceremonies sponsored by the adepts. Candles are 
sent to Beni Ouarad once a month. Money collected at the name-day 
ceremonies is kept by the treasurer, although an adept is given a little 
if he is in dire need. The proceeds from the curing ceremonies are 
saved for the children of the saint, but a part is given to all of the foqra. 
The amount is decided by the muqaddim, the treasurer, and three wit¬ 
nesses. 

A report is made each month, but it is never shown to the mizwar. 
Two days before the annual pilgrimage, the muqaddim, the treasurer, 
the registrar, and three witnesses meet in the zawiya under lock and 
key to decide how much they will give the poorest descendants of Sidi 
Ahmed and how much they will give the mizwar. It appears that each 
zawiya gives a little extra to the poorest of Sidi Ahmed’s children. The 
exact amount is determined on the basis of individual need. (Envelopes 
containing the funds are given during the musem.) The remainder of 
the year’s proceeds are given to the mizwar for distribution. The day 
before the musem, the muqaddim, the treasurer, the registrar, and the 
three witnesses all visit the mizwar at Beni Ouarad and give him the 
money. Before counting it, he gives a token amount (called baraka) to 
the registrar and the witnesses and says a fatha for them. Then he 
counts the money and divides it into four equal parts; he keeps three 
of these and gives the fourth to the muqaddim. The muqaddim divides 
his fourth into three parts; he keeps two and gives the third to the 
treasurer. Thus he receives one sixth of the year’s proceeds, and the 
treasurer one twelfth. The mizwar does not say a fatha for them. 14 

14 In 1968 the Dghughiyyin of Mekncs collected 780 dirhams: 80 dirhams 
were divided among the poor descendants of Sidi Ahmed, and the mizwar received 
the remaining 700 dirhams. He divided nearly all of this into four parts. Each part, 
it is reasonable to assume, was about 170 dirhams. Thus, the muqaddim received 
about 113 dirhams for his year’s work; the treasurer, about 57 dirhams; and the 
mizwar was left with about 510 dirhams to distribute to the 200 or so descendants 
of Sidi Ahmed. Each descendant theoretically received about 2.5 dirhams from 
the Meknes zawiya. (Meknes probably provided considerably more money than 
any other zawiya, with the possible exception of the Fez team.) 



6 


The 

Shantytown 

Teams 


In Aleknes in 1968 there were seven teams of Hamadsha which were 
not affiliated with the zawiyas of the madina and had no permanent 
meeting place. These teams, which were far more active than the urban 
lodges, operated throughout the city and nearby countryside; their 
members occasionally traveled as far as Fez, or even Rabat, to perform 
their ceremonies. Their center of activity was, however, the bidonvilles 
of Meknes, and both the team members and the muhibbin attracted to 
their performances were primarily from these shantytowns. 

The Bidonvilles 

The population of Meknes, as we noted in the preceding chapter, has 
grown rapidly since the arrival of the French. Rural immigrants poured 
into the city seeking work; most often they came from the Gharb and 
from the countryside around Fez and Meknes—areas in which the 
Hamadsha were particularly active. At first they settled in the madina, 
often in the nooks and crannies of Moulay Ismail’s walls or in the cellars 
of his palaces—which were so extensive and so labyrinthine that they 
lived there undetected, in some instances, for years. By the early thir¬ 
ties there was such a housing shortage in the city that the immigrants 
began to settle in vacant lots just outside the city limits. 

Bordj Moulay Omar, the largest of these shantytowns, sprang up in the 
early thirties near El Minzeh, which had been a pleasure palace for one of 
Moulay Ismail’s ministers. It was conveniently located just north of the new 



102 


THE HAMADSHA 


town of Meknes and within a mile of the industrial quarter. By 1935 the 
municipal authorities had begun to show some recognition of the shanty¬ 
town’s problems, particularly those concerning waste disposal. In July 1936 
there were 400 households; in 1952 there were 3,104 (Franchi 1959:258); 
Bordj had then a population of 11,507. In 1960, four years after inde¬ 
pendence, there were nearly 19,000 inhabitants living on 36 hectares. 

The other bidonvilles of Aleknes have had similar histories. The pop¬ 
ulation of Sidi Baba, where most of my shantytown research was con¬ 
ducted, jumped from 5,249 in 1952 to 12,758 in 1960. My own estimate 
for its population in 1968 was between 18,000 and 20,000. It covered 
5.2 hectares. 

The plans of the different shantytowns vary with the lay of the land. 
Sidi Baba and Bordj Moulay Omar are located on relatively flat terrain. 
They have large squares and appear more open than others which are 
built onto steep declivities. Some bidonvilles are considered poorer and 
less desirable than others. Sidi Baba is usually preferred. The larger 
shantytowns are divided into quarters named after some important 
center of activity such as a steam bath, a school, an oven, or a former 
football field. 1 The largest quarter of Sidi Baba is named after the 
Moroccan veterans of World War II and the French-Indochinese wars 
who settled there in the fifties and make up a large segment of the 
bidonville’s population. Although I was not able to find any correlation 
between quarter or street and tribal area, Franchi found that the in¬ 
habitants of some streets or quarters in Bordj came from the same tribe 
or region of Morocco (Franchi 1959:264). 

The houses of the bidonvilles may be made of wattle and daub, clay 
bricks, clapboard, reeds, tar paper, or scrap metal. Few if any of them 
have more than one story, and this gives the bidonvilles a more open 
quality than the madina, with its narrow streets walled in on each side 
by two- and three-story windowless houses. All but the poorest huts 
have an inner court, which is closed on the top with a grill to prevent 
robberies. Perhaps 40% of the houses in Bordj have three or four 
rooms; most of the rest are smaller (Franchi 1959:272). Only a handful 
have windows facing the street; these are covered with bars and 
shutters. 

Although each household forms a separate and independent unit, 
usually consisting of parents and unmarried children and, occasionally, 

1 These quarters have no official status. Their names may, however, appear on 
mailing addresses. 



THE SHANTYTOWN TEAMS 


103 


paternal grandparents, unmarried siblings, or relatives from the coun¬ 
tryside, rooms are sometimes let to strangers. The bidonville households 
are not closed-in and isolated from one another as are those of the 
madina; women, if only by necessity, are not as segregated as in the old 
quarter itself. The fountains at the end of each street, where women 
and children go for water; the ovens, where bread is baked for each 
household; the steam baths; and the local grocery shops are all centers 
of communal activity. They may be considered neighborhood centers. 

Les commer^ants et boutiquiers sont lies a leur clientele qu’ils coudoient 
chaque jour et avec laquelle ils entretiennent d’excellents relations creant 
ainsi une communaute intime parmi tous ces gens. (Franchi 1959:271) 

Although it is possible to speak of an intimate community, une com - 
munite intime , bidonville life is by no means as harmonious and peace¬ 
ful as this phrase suggests. Rather, it is characterized by feelings of mis¬ 
trust and suspicion. Hardly a day goes by without the spectacle of two 
or three women neighbors in a row—most often over their children. 2 
Sometimes these fights become physical and require hospitalization of 
at least one of the participants. Men, too, fight—often brutally—and it 
is considered unsafe to walk the streets at night. There is much talk of 
robbery. 


Urbanization 

Although the bidonvilles have retained the more open quality of the 
countryside, they have lost its traditional controls—ties of kin and 
mutual obligation—in undergoing what Jacques Berque has aptly called 
urbanisation sauvage. The bidonville families have moved to the city 
as individual units. Often they had and still have no relatives in Meknes; 
or, if they do, their relatives may live on opposite sides of the city. 
Friendship among men is not limited to the inhabitants of single neigh¬ 
borhoods or quarters; but women, whose activities arc far more re¬ 
stricted, usually find their friends and enemies in the vicinity of their 
homes. Often a neighborhood will split into two or more factions, 
which will then take sides in the many disputes that occur in such high- 
density populations. 

■The ease with which women enter these tantrums reminds one of the ease 
with which they fall into trance. 



104 


THE HAMADSHA 


The bidonvilles are administered by the municipal authority. They 
are usually headed by a muqaddim, to whom every individual must 
appeal if he has any affairs whatsoever with the government. The mu¬ 
qaddim is responsible to that khalifa (assistant) of the pasha of Meknes 
charged with the particular ward in which the bidonville is located. 
It was not my impression that neighborhoods or quarters formed po¬ 
litical units from which leaders were likely to emerge. Such indigenous 
patterns of leadership are discouraged by the central government. It is 
said that there are more than a few “secret police” in the shantytowns. 
They can usually be spotted by their more expensive shoes, their trench 
coats, and by a style of behavior reminiscent of James Bond, who is a 
very popular movie hero among them. 

The position of the bidonvilles with respect to both city and coun¬ 
tryside is complex, and is reflected in the relationship between the 
Hamadsha of the urban zawiyas and those of the shantytown teams. 
The bidonvilles are not independent social entities. On the one hand, 
they fall under the municipal administration and, on the other hand, 
they are regarded by the inhabitants of the madina (and, to a lesser 
extent, by the inhabitants of the bidonvilles themselves) as outside the 
confines of the city and part of the countryside. Although it is tempt¬ 
ing to consider this as a result of the rural origin of the shantytown 
denizens, it must be noted that in the Mediterranean world the city is 
seen to end at its walls. All that lies outside the walls of the city is 
countryside. 

Jacques Berque (1958:11) has attempted to describe the division be¬ 
tween the two groups, a division he considers, in some way, greater 
than that between the madina and the European quarter. 

Mais la violence du decalage qui oppose, de part et d’autre des murailles, le 
citadin a son compere rural, est excessive. Le langage, les habitudes, l’his- 
toire, le mode de vie, la figure meme et les vetements ne les opposent moins 
l’un a l’autre. Le grand style urbain reste coupe du pays. Et cette discorde 
de style qui reflete une discorde economique et psychologique, l’appauvrit 
et l’expose en cas de crise, a s’etioler. 

Although Berque is writing of Tunisia, his observations are applicable 
to Morocco as well (cf. LeCoeur 1969:118-120). To what extent this 
split affects the self-image and self-esteem of the rural immigrant to 
the city remains an open question. The bidonville inhabitant does joke 
about his rusticity and is critical of urban ways. He is proud of his 
ceremonies, his music and, if he happens to be a member of a popular 



THE SHANTYTOWN TEAMS 


105 


brotherhood, of the endurance of the country folk in performing the 
hadra. He is, nevertheless, conscious of his poverty, his illiteracy, his 
lack of skilled profession, and the misery of his household. All of the 
shantytown dwellers whom I questioned assured me that they would 
prefer to live in the madina, which was cleaner, more peaceful, quiet, 
and nearer their marketplace and place of work. One woman added, 
“where the people are educated and more civilized.” 

The integration of the bidonville and madina is, nevertheless, to some 
extent facilitated by the very structure of Middle Eastern cities like 
Meknes. These cities are themselves composed of quasi-independent 
units, or quarters, often of distinct ethnicity, which are oriented toward 
the central marketplace and administrative center. Coon (1964:232) has 
employed a cytological simile for describing the city: 

... It might be said that the city itself has a cell-like structure with its 
own nucleus, its own mechanisms for withstanding shocks from within and 
without, and its own ways of absorbing and sloughing off foreign materials. 

The bidonvilles show some signs of becoming quarters of the city, but 
they are, at the same time, developing and maintaining their own sepa¬ 
rate character. This is best illustrated by the recent appearance of a cult 
and musem to Sidi Ma'sud, who is buried in Bordj Moulay Omar. In 
1968 there was some talk of a musem to another, as yet legendless, saint, 
Sidi Baba, whose burial place has not yet been discovered but is be¬ 
lieved to be in the vicinity of the bidonville named after him. 

The principal tie between bidonville and city is economic. Franchi 
(1959:267-268) found that in the late fifties only 14% of the working 
population of Bordj Moulay Omar worked in Bordj itself. Only 30% 
of the total population of the bidonville (54.8% of the potential work¬ 
ing population) were employed then, 63% of these as day laborers. 
Although there have been a number of changes in the economic struc¬ 
ture of Meknes since Franchi’s study—it is possible that the percentage 
of employed has decreased since 1958—it seems that the majority of 
workers in Bordj are still journeymen and earn well under five dirhams 
a day. Because of the large number of veterans who receive pensions 
from the French government, the situation in Sidi Baba is more com¬ 
plex. Nevertheless, the majority of workers there are laborers who do 
not work locally. This pattern also appears to hold for the other bidon¬ 
villes. 

The shantytowns are also tied to the madina commercially. Although 



106 


THE HAMADSHA 


the larger ones have little vegetable and meat markets as well as grocery 
shops, many of the inhabitants of the bidonvilles prefer to do their 
marketing in the madina, where prices are lower. The bidonville mar¬ 
kets, which maintain a very small inventory, are stocked each day, ex¬ 
cept Wednesdays and Saturdays, from the madina markets themselves. 
Wednesdays and Saturdays are the days of the “wholesale market” just 
outside the city limit. The mark-up in the bidonville is considerable. 
There are a few small shops in the shantytown where one can buy 
tobacco or mats or have clothes mended or shoes repaired; however, it 
is impossible to buy most standard household goods in the bidonvilles. 
At least one member of the average shantytown household—even those 
in which no one works in the city—visits the madina once a day. There 
is a bus service from Bordj Moulay Omar, but from Sidi Baba it is 
necessary either to walk, bicycle, or take a taxi. 

Although the average bidonville inhabitant is tied to the madina eco¬ 
nomically, he is not integrated into its social life. As a newcomer to 
the city, he has in most instances married a woman from his own 
family or tribe. Second marriages—after divorce or death—are made 
more often with other bidonville inhabitants in a similar situation. 
Children who have been bom and educated in Meknes are now begin¬ 
ning to marry other inhabitants of the shantytowns, regardless of tribal 
or family affiliation. Marriages between bidonville dwellers and mem¬ 
bers of old madina families are still rare. Marriage ceremonies are not 
those of the city but of the country. Patterns of friendship follow the 
same lines as marriage. Tribal and occupational relationships are the 
foremost determinant. 

The “religious ties” with the madina are the most difficult to deter¬ 
mine. Although the larger bidonvilles have their own mosques—often 
shacks with clapboard minarets—those men who pray regularly prefer 
to attend the madina mosques whenever they can, and especially on 
Fridays. They seldom belong to any of the old-quarter zawiyas; if they 
have an affiliation with a brotherhood, it is with one in the bidonville. 
Indeed, there is considerable hostility between urban and slum brother¬ 
hoods. Saints’ tombs in the madina, especially Sheikh al-Kamal, are 
visited regularly by both men and women, although the majority of 
bidonville inhabitants seem to have greater faith in Sidi e Ali and Sidi 
Ahmed. They visit and often invoke the saint of their countryside, by 
whom they are frequently msannad. The square in front of Sheikh 
al-Kamal is filled each Friday afternoon with inhabitants of the shanty¬ 
towns who have come to attend the hadra of the Hamadsha, the Mili- 



THE SHANTYTOWN TEAMS 


107 


ana, the Jilala, and the ffsawa, or to listen to storytellers, musicians, holy 
fools ( majdubin ), and the like. 

The Shantytown Teams 

The Hamadsha of the shantytowns differ from those of the madina 
zawiyas, as would be expected, in terms of their beliefs, their ceremonial 
style, and their organization. They may be divided into two groups: the 
team members and the muhibbin. The team members are professional 
Hamadsha—they earn their living by performing the hadra—and appear 
to be much more explicitly concerned with the jnun and with sickness 
than the adepts of the madina, who are primarily concerned with the 
saints. Their ceremonies are simpler—they do not recite litanies—and 
more violent and spectacular. The public ones fall midway between 
spectacle ( fraja ) and religious ceremony ( krama y sadaqa). The private 
ceremonies are primarily curing ceremonies and not homages to the 
saints. 

Not only are the performances designed—often crudely—to maxi¬ 
mize profits, but the team members actively seek out muhibbin and 
threaten them with illness if they do not sponsor a hadra. Usually these 
threats arc made in a joking, offhand way, but they leave their mark on 
a group of people who are highly susceptible to such suggestions. The 
bidonville Hamadsha are “exploiters,” and they are seen as such not 
only by the descendants of Sidi c Ali and Sidi Ahmed and the foqra of 
the madina but also by their “victims,” the muhibbin themselves, who 
stoically accept their dependence as their fate. The devotees conceive 
of their attraction to the Hamadsha as the will of ‘Aisha Qandisha and 
the other jnun, whose ways are neither to be questioned nor under¬ 
stood. Like the team members, they are concerned primarily with the 
jnun, who constantly interfere with their lives, and only secondarily 
with the saints, who are seen as providing the means ( baraka) for man¬ 
aging the jnun. Baraka is not conceived in its more abstract quality as 
plenitude, well-being, and completion, but in its more concrete sense 
of material well-being, good health, and luck in everyday affairs. The 
devotees perform the hadra to obtain something, and are not at all un¬ 
willing to pay for it—even to men they recognize as charlatans, quacks, 
whoremongers, and money-grubbers. 

The seven teams are dedicated to either Sidi c Ali ben Hamdush or 
Sidi Ahmed Dghughi. The allegiance of the members, however, may be 
mixed. The teams usually have seven members, although there is tech- 



108 


THE HAMADSHA 


nically no limit to membership or particular virtue in this number. In¬ 
deed, the number seven is popularly considered unlucky. The teams are 
not as strictly and as formally organized as the zawiyas, and on occa¬ 
sion—especially during the public performances on Fridays and holi¬ 
days—two or more teams will combine. Often, too, individual mem¬ 
bers of a team which is not performing on a given night will join the 
hadra of another team. They may receive a small gift for this, but their 
unexpected arrival is usually regarded as an intrusion and is resented. 
There is considerable friction between the different teams. 

Each of the bidonville teams is headed by a muqaddim who, like the 
muqaddim of a zawiya, receives a letter of charter from the mizwar 
of the saint he follows and carries the saint’s flag. He is charged with 
the administration of the team, the scheduling and organization of cere¬ 
monies, and the division of proceeds; he acts as a dance leader during 
the hadra and cares for the sick. The mizwar’s control over the muqad¬ 
dim of a bidonville team is not as great as his control over the head of 
an urban zawiya. He does not choose, or even confirm, the election of 
the muqaddim. Usually the muqaddim was a team member or an active 
devotee who felt that he had enough baraka to attract a following and 
who conducted a few trial performances before presenting his case to 
the mizwar. Often he replaces an old or deceased muqaddim. The miz¬ 
war grants the new muqaddim a letter unless there is active opposition 
from other teams in the area. 

Both mizwars hold the bidonville Hamadsha in some contempt, con¬ 
sider them cynically as a good source of income, and term their letters 
to them acts of charity graciously given to help a group of poor 
wretches earn a living. The letter is the mizwar’s only hold over the 
muqaddim; the muqaddim must then present it to the municipal gov¬ 
ernment, which grants the team permission to perform its lucrative 
public performances. 

The other members of the team are musicians and dancers. There is 
usually one snare drummer ( tabbal ), who is familiar with the words of 
the various trance-inducing musical phrases (ribs), three gwal players, 
and two ghiyyata who also play the recorder (nira) and the guitar 
( ganbri ). Each muqaddim tries to have at least one sharif —descendant 
of the Prophet—in his team, because a sharif, who is endowed with 
baraka, attracts a following and will bring in more money. The sharif 
may play an instrument; he usually aids the muqaddim in caring for 
the sick. 

There is no formal initiation: “The muqaddim notices that you are a 
good muhibb and invites you when one of the regular team members is 



THE SHANTYTOWN TEAMS 


109 


unable to make it. Or he may ask you to dance because he has seen that 
you are a good dancer.” If the novice is young, he is left with all the 
dirty work of the team—running errands, carrying the instruments, 
and informing team members of a ceremony—and is the object of a 
great deal of teasing. He may be required to shave his head, leaving 
only a small pigtail. The Hamadsha of Meknes do not have a single hair 
style, and the novice will usually let his hair grow long again. 

The ghita players are considered part of the team, even though they 
receive greater compensation than the other musicians and perform at 
other ceremonies as well. They do not, however, play for more than 
one bidonville Hamadsha team. The difference in status between the 
ghita players of the zawiyas and those of the teams must be understood 
in terms of the division of wealth. The foqra do not divide the proceeds 
of a ceremony among themselves; theoretically, only the poorest of 
them can expect to receive a part of the small portion of total proceeds 
allotted to them. This allotment is referred to as alms ( sadaqa ), or 
baraka, or bread, a symbol for baraka, and is considered as an act of 
charity. The team members, however, divide all of the proceeds among 
themselves and do not normally consider their part alms. 

The importance of the ghiyyata, who control the progress of the 
ceremony by playing the appropriate rihs, often undermines the posi¬ 
tion of the muqaddim, whose authority, aside from what personal 
charisma and esoteric knowledge he may have, lies with his possession 
of the mizwar’s letter. The teams tend, therefore, to be very fragile. 
The members of the seven teams operative in the bidonvilles of Meknes 
in 1968 were constantly shifting their allegiance from one muqaddim 
to another. Whatever stability there was resulted from the economic 
success or failure of a particular team. Ghiyyata perform at marriages 
and circumcisions; and during the period immediately following the 
harvest, when the majority of marriages and circumcisions take place, 
the Hamadsha seldom perform the hadra because they are unable to 
find ghiyyata for the occasion. This is the period of the greatest tension 
within the teams. They often break up, to be reintegrated later in the 
year. 

The position of the ghiyyata in the larger social setting must be dis¬ 
tinguished from that of the other team members. They form an occu¬ 
pational group and, as ghiyyata, are treated with respect. The majority 
of them come from illiterate, peasant Gharbawi families which usually 
included some Hamadsha. The fathers of the majority of them were 
not ghiyyata. The musicians learned to play the ghita from a m c alleni, 
or master, in their home villages, who may have noticed their musical 



110 


THE HAMADSHA 


ability and taken an interest in them, teaching them first to play the nira 
and then the ghita and ganbri. Most of them, much to their parents’ 
disapproval, then joined a troupe of wandering musicians ( sheikhat )— 
these travel from village to village, usually accompanied by a prostitute 
or two, performing at name-day celebrations, circumcisions, marriages, 
and other festivities—before coming to Meknes. Although many of the 
ghiyyata were familiar with the music of the Hamadsha, as well as that 
of the 'Isawa, the Jilala, and the Gnawa, they were seldom affiliated 
with a team of Hamadsha before their arrival in the city. Usually, they 
joined one shortly afterward. The team provides them not only with 
a means of earning a living but also with the chance to form contacts 
with families who would later invite them to perform at family cele¬ 
brations. They regard themselves and are regarded by others as pro¬ 
fessional musicians who are members of the guild of all ghiyyata, in¬ 
cluding those not attached to any brotherhood. They are proud of their 
profession. 

The other team members do not form as homogeneous a group. Like 
the ghita players, they are younger than the foqra of the madina and 
have almost all been bom in the countryside, where they performed 
the hadra. Like many of the madina Hamadsha, they began to dance 
in their early teens. They too did not become team members until 
coming to Meknes. The motive was usually financial. The majority of 
them come from areas in which the Hamadsha were active, and many 
of them report that other members of their families perform the hadra. 
It is not unusual to find families, especially among the head-slashers, in 
which every member of the immediate family has performed the hadra. 
But it is also not unusual to find team members who report that no one 
in their immediate or more distant families has danced. The majority 
of the members I interviewed came from the Gharb or from the region 
around Fez and Meknes; very few come from the Zerhoun. Most of 
them are illiterate and, as they put it, have no other “occupation” beside 
the hadra. They will take odd jobs whenever they can and will work 
the harvests, preferably around their old homes. Those who work have 
menial jobs—selling doughnuts or candies or working as garbage col¬ 
lectors or street cleaners—or make a little money by preparing magical 
amulets and formulae, predicting the future, or giving advice to the 
jinn-possessed, the jinn-struck, or those suffering from the evil eye 
or a magical curse. A few of them—they usually come from the Wulad 
Khalifa in the Gharb—claim to be descendants of the Prophet and earn 
some money reciting fathas, massaging aching limbs, or spitting into 



THE SHANTYTOWN TEAMS 


111 


water which then, filled as it is with their baraka, is drunk by the ailing 
client. 

There is considerable ambivalence in the attitude of the average 
bidonville inhabitant toward them. Although the shantytown dwellers 
regard them with a certain awe and treat them, in face-to-face en¬ 
counters, with respect, the team members are never as respected and as 
admired as the ghiyyata (or the foqra of the zawiyas). In private, they 
are often referred to quite disparagingly. Most of the bidonville dwell¬ 
ers, including the team members themselves, do not want their children 
to become team members or even muhibbin. There is little prestige 
associated with their role, although within their own ranks there ap¬ 
pears to be a hierarchy of prestige. Those who abandon themselves 
most dramatically to excesses of self-mutilation and those who have 
access to the jnun have the most prestige. They are all treated with 
great condescension by the educated of the madina, and the foqra con¬ 
sider them “worthless Hamadsha.” 

The Devotees 

The vast majority of the bidonville Hamadsha are devotees ( muhib¬ 
bin) \ and like the devotees of the city, they are difficult to trace so¬ 
cially. They may be scaled with respect to intensity of involvement or 
frequency of performance in the same manner as the madina devotees. 
The most intensely involved—those who fit into the first 12 categories 
of Figure 2—have the most prestige; they form networks which 
are centered around a particular team member. Usually, this is a muqad- 
dim or ghita player. He invites them whenever he gives a performance, 
especially if the performance is near their homes. Often the members 
of the network come from the same tribe or region as the muqaddim or 
ghita player; however, one muqaddim, who had a very large following, 
came from a region of Morocco which had produced very few immi¬ 
grants to Meknes. 

Most significant is the fact that the devotees of the bidonvilles are 
centered around particular team members whom they follow. The pat¬ 
tern is in this respect different from that of the madina. The team mem¬ 
bers, who are constantly sponsoring ceremonies themselves, play a 
much more active role in inviting the muhibbin to ceremonies than the 
foqra do. There is an incipient, albeit restricted, form of leadership. 
Were the teams themselves more stable , one might expect to find a 
rudimentary association of muhibbin around each team. This is not the 
case. One reason for the more active role of the team members in the 



112 


THE HAMADSHA 


invitations is economic. The more devotees present at a ceremony, the 
more money the team can hope to collect and use for its own devices. 
I do not mean to imply that the foqra are not sufficiently motivated to 
maximize their profits, but simply that there is a greater degree of 
urgency in earning money when one’s very existence depends on it. 
A bidonville ceremony is much more a business affair than a ceremony 
in the madina. 

Some of the muhibbin of the bidonville teams do not live in the 
shantytowns but in the madina itself and in the villages in the vicinity 
of Meknes. Although devotees who live in the madina arc, like their 
confreres in the shantytowns, usually newcomers to Meknes, there are 
a few who come from more established families. Unlike the devotees 
of the shantytowns, they usually have affiliations with one of the 
Hamadsha zawiyas of the old quarter; but, like Boualem, they require 
the hadra more often than the zawiyas perform. There is, in short, no 
complete division between the muhibbin of the madina and those of 
the bidonvilles. Although they tend to split into two groups along an 
urban-bidonville (rural) axis, the split is not complete. They do share 
their devotion to the saints and e Aisha Qandisha, and their performance 
of the hadra. 

There are many more female muhibbin in the bidonvilles than there 
are among the zawiya-affiliated devotees; and the women, who are not 
as restricted as women in the old town, play a more active part in the 
ceremonies. The wives of most, but not all, of the team members also 
perform the hadra; this cannot be said for the wives of the male de¬ 
votees. One often finds a devotee whose wife not only is not a devotee 
but disapproves of the Hamadsha altogether. Similarly, there are mu¬ 
hibbin with husbands who are not only not devotees but who tolerate 
their wives’ performances with the usual superiority of an Arab male 
toward an Arab woman. Spouses of either sex, however, frequently re¬ 
port becoming devotees after marrying a devotee. 

The average male muhibb resembles the average bidonville inhab¬ 
itant. He is illiterate, an unskilled laborer of peasant background who 
may or may not have full employment and who leaves the city to work 
the harvests in the spring and summer. He is a newcomer to Meknes 
and was born in a part of the countryside where the Hamadsha were 
active: the Gharb and the regions of Fez and Meknes are the most 
common. I have found far fewer from the Zerhoun than I did in the 
madina. This may be the result of continued contact between the Zer¬ 
houn population and Meknes, which has provided the Zrahna with 
“better contact” in the madina. Like the muhibbin of the cities and the 



THE SHANTYTOWN TEAMS 


113 


foqra and team members, the devotees of the shantytowns usually re¬ 
port at least one other family member who performs the hadra. Like 
the other Hamadsha, they usually started to perform the hadra in their 
teens after a sickness—most often paralysis—which was interpreted as 
an attack by the jnun. From that time on they have felt the need to 
dance, and they suffer from depression, general malaise, and pinching 
bones if they do not dance for a prolonged period of time. Sometimes 
they become paralyzed again. Their first performance usually preceded 
their arrival in Meknes, 


Finances 

The economic organization of the bidonville Hamadsha is different 
from that of the zawiyas. Unlike the foqra, who “work” for their saint, 
the team members of the bidonvillcs work for themselves. Their public 
and private ceremonies are always considered curing ceremonies, and 
they do not recite litanies or attend name-day celebrations. Just after 
each performance, they divide all the money and gifts—if there should 
happen to be any—among themselves. Nothing is left for the saint. 
During the weeks immediately preceding the musem, however, a part 
of the proceeds from each performance is put aside for the saint. Often, 
at this time, the teams will go from house to house begging alms for 
their saint. At the musem, this money is given to the mizwar, who re¬ 
news their letter of charter without saying a fatha over them. The 
whole exchange is conducted like a business transaction with much 
bargaining, complaining, and cursing, and it does not have the sacro¬ 
sanct quality of a zawiya’s presentation of proceeds to its mizwar. The 
team members are simply renewing their contract. 3 

* During the musem of 1967, one team was not able to give the mizwar of Sidi 
c Ali as much as was their custom (125-150 dirhams) because K., the muqaddim, 
fell sick shortly before the musem, and his team was unable to perform for sev¬ 
eral months. The mizwar of Sidi c Ali did not consider this to be an excuse and 
threatened to withdraw his letter and inform his relatives in the municipal gov¬ 
ernment of the change in K.’s team’s status. When K. still refused to pay, the 
mizwar cursed him and called K.’s wife a pagan, a very strong insult. He did not 
in fact withdraw the letter, but sent his son to K.’s house several times during the 
year to collect as much money as he could. K. threatened to switch allegiance to 
Sidi Ahmed, but was urged not to “abandon his saint” by the other team members. 
K. ended up by giving the mizwar’s son about 10 dirhams at each visit—40 or 50 
for the year. In 1968 K.’s team was more active and was able to collect enough 
money (160 dirhams) to satisfy the mizwar. K. and the other team members do 
not hesitate to curse the mizwar and his son—not the saint; and K. who is reputed 
to have great knowledge of magic, claims to have made much evil magic against 
the mizwar, his son, and his brother. 



7 


The 

Circle of 
Exchange 


The three levels of Hamadsha organization considered in the last three 
chapters are interrelated in a way that is typical of all levels of Ha¬ 
madsha organization and that receives ritual expression during the an¬ 
nual pilgrimage to Beni Rachid and Beni Ouarad. They form an ex¬ 
emplary complex, the analysis of which provides an understanding of 
the “internal workings” of the organization as a whole and of its rela¬ 
tion to the larger social scene. Only those aspects of the “external rela¬ 
tions” of the Hamadsha bearing on the therapeutic process are con¬ 
sidered here. 

The Hamadsha organization under consideration may be sum¬ 
marized as shown in Table 3. 

The Hamadsha are seen to fall into tw o structurally parallel organiza¬ 
tions—the descendants and followers of Sidi c Ali and the descendants 
and followers of Sidi Ahmed—between which there are certain ritual¬ 
ized ties. Both organizations are divided into two distinct levels: the 
wulad styyid, who form a bounded group with explicit rules of mem¬ 
bership—agnatic descent from the ancestral saint—and the adepts 
(foqra ) and devotees ( muhibbin ) of the brotherhood, divided into a 
series of teams whose organization varies from town to country. (The 
bidonville teams tend to follow the organization of the countryside.) 

Membership in the teams is achieved. There are no explicit rules for 
recruitment. Each team is surrounded by a number of devotees, who 
may be scaled with respect to the intensity of their involvement. 
Acknowledgment of one’s devotion to the Hamadsha saints and to 
their cult is the only criterion for being considered a muhibb. Al- 



THE CIRCLE OF EXCHANGE 


115 


TABLE 3 



Sidi *Ali 

Sidi Ahmed 

Zerboun 

Mizwar 

Wulad Siyyid 
(Ta'fa) 

Mizwar 

Wulad Siyyid 
(Taya') 

Madina 

Bidonville 

Madina 

Bidonville 

Zawiya 

Taifa 

Zawiya 

Tatfa 

(foqra) 

(team member) 

(foqra) 

(team member) 

(ghiyyata) 

muhibbin 

muhibbin 

(ghiyyata) 

muhibbin 

muhibbin 


though the various teams are distinct from one another, the devotees 
are often affiliated with more than one group. Often, one finds a 
muhibb who is connected not only to the zawiya of one saint but also 
to one or more bidonville teams, and who does not hesitate to attend 
the performances of the other zawiya. 

The mizwar, who may be considered the direct representative of the 
saint, is in complete charge not only of the descendants but of the 
followers of his ancestral saint. He is the nodal point of the organi¬ 
zation, and his importance to the organization is ritually reconfirmed 
every year at the annual pilgrimage. 


The Musem 

Each year, on the sixth and seventh days after the Prophet’s birth¬ 
day, or midud, the component organizations of the Hamadsha complex 
come together for the musem at Beni Ouarad and Beni Rachid. Tens 
of thousands of pilgrims—foqra, team members, muhibbin, as well as 
the curious, the pious, and the bored—from all over Morocco gather 
together to pay homage to Sidi e Ali and Sidi Ahmed. 1 

On the days immediately preceding the musem, pilgrims arrive and 
set up tents in the village of the saint to whom they feel closest. 
Traveling vendors, entertainers, hawkers, and pickpockets who make 
the round of pilgrimages each year also arrive to ply their trade. Team 
members settle down in groups and begin to perform the hadra almost 

1 1 estimate that there were between 35,000 and 50,000 pilgrims at the museum in 
1968. This was also the estimate of the chief of the Gendarmerie Royale at Moulay 
Idriss. Schoen (1937:91) estimated 35,000 for the musem in 1935. 




116 


THE HAMADSHA 


immediately on arrival. Their dancing continues intermittently for 
three or four days, until the musem ends. The foqra of Meknes and 
Fez, as well as the leaders of other important urban lodges, are given 
special housing by their mizwar. It is at this time that the various tai'fas 
bring the year’s proceeds to the mizwar and make the rounds of other 
saint’s descendants to whom they feel especially bound. A constant 
stream of pilgrims visits the two saints’ tombs, the baths at Beni Rachid, 
and the shrines to c Ai'sha Qandisha, leaving a few dirhams, a candle or 
two, or some other gift at every stop. (The mizwars usually grant 
charge of the baths and the shrines as concessions to the highest bidders 
in the villages a few days before the musem.) Numerous chickens are 
sacrificed along with the rams, he-goats, and bulls that are usually— 
and traditionally—brought by the urban zawiyas . 2 Each year one of 
these zawiyas brings a special cover for the catafalque and parades up 
and down the road with it. 

The musem of Sidi Ahmed takes place on the sixth day after the 
Prophet’s birthday. All morning and afternoon, teams from all over 
Morocco dance their way—in trance and often slashing their heads— 
down the narrow main street of Beni Ouarad to the saint’s tomb; and 
there, in front of the mausoleum, they dance even more violently, some 
of them slashing with a halbard ( shaqria ). The mizwar and the other 
saint’s descendants do not participate in this part of the pilgrimage; 
they stay in their houses, entertaining important guests. 

The mizwar of Sidi c Ali is invited to dine with the mizwar of Sidi 
Ahmed, if relations between the two permit this. (The musem is one 
of the rare occasions on which the mizwar of Sidi c Ali will allow him¬ 
self to be seen in Beni Ouarad.) Toward sundown, when all the teams 
have paid a visit to the saint’s tomb and all the pilgrims are anxiously 
awaiting his appearance, the mizwar leaves his house, mounts a white 
stallion , 3 and parades down the main street of Beni Ouarad to the tomb 
of Sidi Mejma c at the opposite end of town . 4 Bread and spoonfuls of 
couscous—both said to contain baraka—are given to the pilgrims just 
before his appearance. He pays homage to Sidi Mejma e by circum¬ 
ambulating the tomb and then returns to his ancestor’s mausoleum, 

“For the musem of 1968 the mizwar of Sidi Ahmed required 4 cows, 20 sheep, 
between 300 and 400 chickens, 200 kilograms of honey, 100 kilograms of butter, 
800 kilograms of couscous, and over a ton of flour. I have no figures for tea, 
sugar, coffee, and mint. It is said that the mizwar does not collect enough during 
the musem to pay for all this. 

8 It is explained that e Ali, the Prophet’s son-in-law, rode a white stallion. 

4 1 know of no legends about Sidi Mejma c or his relationship to Sidi Ahmed. 



THE CIRCLE OF EXCHANGE 


117 


which he is supposed to visit but is seldom able to reach because of the 
crowds. He is followed by various teams of Dghughiyyin, who per¬ 
form the hadra in tight separate circles ( halqat ). The foqra of Fez, 
whose muqaddim is related to the mizwar by marriage, follow im¬ 
mediately behind him. The foqra of Meknes end the procession, ac¬ 
companied by two or three men on horseback playing a long brass 
(or copper) horn called a mfa c . 6 

On the seventh day after the Prophet’s birthday, much the same 
sequence of events is repeated at Beni Rachid. The mizwar of Sidi 
Ahmed is asked to lunch, but then he returns to his village, mounts his 
stallion, and rides back down to Beni Rachid, followed by thousands 
of pilgrims. The Dghughiyyin perform the hadra. Traditionally, the 
mizwar followed a path through the olive groves between the two 
villages and stopped at the two palm trees, to the north of Beni Rachid, 
to be greeted by the mizwar—also on horseback—of Sidi c Ali. The 
mizwar of Sidi Ahmed would then dismount and, on foot, follow the 
other mizwar—still on horseback—to Sidi c Ali’s tomb and back . 6 In 
recent years, however, the mizwar of Sidi Ahmed has refused to follow 
the traditional path, preferring the tarred Zerhoun road. He no longer 
enters the village. Instead he stops just short of the palm trees, waits 
for the mizwar of Sidi £ Ali, greets him perfunctorily, and immediately 
rides back to his own village. The mizwar of Sidi c Ali returns to his 
house without visiting his ancestor’s tomb. Thousands of tired pilgrims 
who have looked forward to this meeting for several days now pack up 
their belongings and prepare to leave that night or the following 
morning. 

The musem, which is part of a complex cycle of pilgrimages to 
various saints’ tombs in northern Morocco, is an amorphous affair with 
many different functions. It may be said to consolidate the various 
components of the Hamadsha complex by reaffirming their position 
within the Hamadsha scheme of things and by justifying the exchange 
of wealth and baraka which takes place both at this time and through¬ 
out the year. Traditionally, it reconfirmed the relative status of Sidi 
c Ali and Sidi Ahmed and their children and followers. Today, it pro- 

8 The nafa c is rarely used at Hamadsha ceremonies. I once saw it played by the 
Dghughiyyin at a circumcision ceremony for their mizwar’s son. It is often 
played at circumcisions. 

•The dcscendents of Sidi 'Ali insist that this is as it should be. The leader of 
the descendants and followers of Sidi 'Ali’s “slave” should follow Sidi £ Ali’s most 
important descendant afoot. 



118 


THE HAMADSHA 


vides a vehicle for the expression of tensions that exist between the two 
economically vying groups. The mizwar’s position, as representative of 
his ancestor and as leader both of his lineage and of the brotherhood 
devoted to his saintly ancestor, is reaffirmed symbolically—rein¬ 
forced behaviorally, one might say, given the heightened suggestibility 
of the exhausted and often entranced pilgrims. His hospitality, which 
is symbolized by the distributing of spoonfuls of couscous and hand¬ 
fuls of bread from his kitchen, demonstrates to the pilgrims a mag¬ 
nanimity and generosity commensurate with a vast endowment of 
baraka. 


Baraka and the Exchange of Wealth 

It must be remembered that bread is also a symbol of baraka, and 
baraka is one element in the ideal exchange system that unites the 
various levels of Hamadsha organization. The muhibbin (and general 
population) give money and gifts—wealth—to the adepts of the order 
and to the saint—by extension, to his children. In return for their gifts, 
the donors obtain baraka. The adepts who have worked for the saint 
give what they have collected to the saint—via the mizwar. In return, 
they receive baraka. The mizwar, as representative of the saint and his 
descendants, distributes the wealth he has received from the adepts and 
from the muhibbin directly to the children of the saint (including him¬ 
self), who are considered to be poor. 7 

It is incumbent upon the mizwar and the children of the saint to 
provide for individual pilgrims to the shrine and for pilgrims at the 
annual musem. The system of exchange is, then, a system of redistribu¬ 
tion of wealth through a series of transfer agents from the muhibbin 
who have wealth to those who do not, the poor and ailing who visit the 
shrine. This redistributive system is schematically represented in Figure 
5. The thin lines indicate the direction of flow of wealth and the thick 
ones that of baraka. 

The fact that the wulad siyyid and the foqra are considered to be 
poor —foqra is derived from the Arabic for poor—enables them to reap 
some of the benefits of the alms they receive. This is not considered a 
salary, or even as their due. The gifts are received as a blessing and are 
called baraka. It is noteworthy that the gifts to the “poor foqra” are 
given before the proceeds of any one night are counted, and that the 


7 Contrast Gellner 1969:74 et seq. 



THE CIRCLE OF EXCHANGE 


119 


PILGRIMS 

(POOR AND AILING) 

t I 

WULAD SAYYID 


t t 



GENERAL POPULATION 

FIGURE 5. 

mizwar gives a gift to the three witnesses who accompany the 
muqaddim and other officials of the zawiya to his house with the year’s 
proceeds before he counts them. He says a fatha for the witnesses but 
not for the muqaddim, the treasurer, and other officials. His gift is re¬ 
ferred to as baraka, or bread. 

A detailed examination of the “logic” of baraka 8 is necessary for an 
understanding of this system of exchange. Baraka, as Geertz has 
pointed out, may be an attribute of an individual or may reside in an 
object. Westermarck and others have looked to the object itself for 
reasons which qualify it as endowed with baraka. They have failed, 
however, to consider the object as symbolic of a group or of some 
social or psychological process. It is indeed possible, and even likely, 
that the extraordinary quality of the object renders it a fit receptacle 


* Here I refer to baraka within the Hamadsha context only. The term is widely 
used in Morocco for many different things, all of which are related in differing 
degrees to Geertz’ gloss. 



120 


THE HAMADSHA 


for baraka, but this does not necessarily enhance our understanding of 
the logic of baraka and its relation to social and psychological proces¬ 
ses. These objects—sand, water, henna, and the like—which the Ham- 
adsha believe to contain baraka are in fact extensions of the baraka of 
the saint. They possess no unique or extraordinary quality in them¬ 
selves, and need not concern us here. 

As an attribute of human beings, baraka may be either institutional¬ 
ized or personal. The former is passed agnatically; the latter is a 
personal endowment which is not heritable, though it may be con¬ 
tagious. Among the Hamadsha, the descendants of the saints are en¬ 
dowed with institutionalized baraka. They are bom with it and defined 
by it. In this sense, baraka is symbolic of the agnatic principle of 
descent—the rules of group membership. It does not reflect any partic¬ 
ular quality in the saint’s descendants, nor does it provide them with 
good fortune, good health, or any of the other more or less concrete 
benefits associated with baraka. It is contagious. 

The baraka of the foqra and team members is personal. Here we 
must be careful to distinguish between the baraka an adept obtains 
from the saint in the same manner as a pilgrim—baraka which is trans¬ 
lated into some concrete potential or actual state of being—and the 
baraka he possesses as a follower of the saint. The latter he cannot 
obtain by his own devices. It is descended from God via the saint. It 
is contagious. Possession of it may be said to define the group. In this 
case, working for the saint—performing the hadra—is the principle of 
membership. 

Baraka is always ascribed to someone or something other than self. 
It is always possible to refer to someone or something as endowed 
with baraka, but it is impossible to refer to oneself as possessed of 
baraka. A man who is endowed with institutionalized baraka might 
make such a statement, but this would be considered gross immodesty 
and almost never happens. A man who is endowed with personal baraka 
is equally incapable of making such a claim. In fact, were he to state 
that he possessed baraka, his endowment might well vanish. 

Baraka is also contagious. It may be passed from an endowed object 
or person to a supplicant by a number of ritual practices. The moment 
the baraka has passed to the supplicant—except in extraordinary cir¬ 
cumstances like Sidi 'Ali’s ingestion of Bu'abid Sharqi’s vomit—it is 
no longer contagious. It ceases to be itself. The individual to whom 
the baraka has been passed obtains something—the state or potential 
state of good health, good fortune, business success, or fertility. A 



THE CIRCLE OF EXCHANGE 


121 


transformation analogous to the conversion of energy into matter, of 
semen into infant, 9 has occurred. Baraka is thus a potentializing force 
which in the process of transference is actualized into that which is 
sought. It should be noted that the specific desire for the baraka must 
precede its transfer. It is impossible under normal circumstances to 
hoard it and then decide upon its use. 

In the Hamadsha scheme, baraka is passed from the saint—the in¬ 
exhaustible source of baraka—through a series of transfer agents: the 
wulad siyyid, who are extensions of the saint himself and who have 
received their endowment biologically (through the semen of their 
ancestor), and the foqra, whose endowment is weaker than that of the 
saints’ descendants, and must be continually renewed through work. 
The saint, his descendants, his followers, and the ritual paraphernalia 
associated with him, all possess baraka which is passed to those who 
do not possess it in exchange for wealth. The passage of baraka mir¬ 
rors—and, in a sense, potentializes—the passage of wealth in a redis¬ 
tributive system which serves ideally to care for the poor and ailing 
and is congruent with a fundamental tenet of Islam: charity. It is seen 
to reside in precisely those ideologically selfless individuals, living in 
self-imposed poverty, who act as transfer agents in the redistributive 
system. 

The ideal exchange of baraka and wealth is not, however, consistent 
with what I was told, or with what I observed. 10 What I observed is 
this: The nuhibbin were the donors of wealth and the recipients of 
nontransferable baraka. They would either give directly to the saint 
or to the adepts of the order. They would always speak of receiving 
baraka from the saint if they left something at his tomb. They usually 
referred to receiving baraka when they gave to the adepts of the 
madina; when they gave something to the bidonviile Hamadsha, they 
were generally more concrete in their demands and, or so it seemed 
to me, less likely to refer to baraka. The exception was when the 
adepts of the bidonviile were in trance. Then they would refer to 
baraka. While many of the devotees referred to the baraka of the 
foqra, they seldom referred to the baraka of the team members. The 

# The passage of baraka is in many ways analogous to the sexual act. 

10 The difference between what all flamadsha expected to happen ideally and 
what in fact happened—that is, between the normative statement of the exchange 
system and a realistic one—created a dissonant situation to which the Hamadsha 
reacted, as we shall see, in accord with their vested interests and their structural 
position within the flamadsha complex. The explanation offered in the text is 
indebted to Festinger’s (1956) theory of cognitive dissonance. 



122 


THE HAMADSHA 


foqra kept no money except the little that they divided among their 
poorest members; actually, each of them received a small share. These 
allotments were called bread, or baraka. In fact, the majority of the 
proceeds were given to the mizwar. 

The team members, on the other hand, kept the majority of pro¬ 
ceeds and gave the mizwar the bare minimum necessary for the preser¬ 
vation of their team. The mizwars—and here I am simplifying—kept 
for themselves as much of the wealth from both the saint’s tomb and 
the brotherhoods as they could. What wealth they distributed to the 
other wulad siyyid fell into the hands of the most powerful and influ¬ 
ential. Neither the mizwar or the other saints’ descendants cared for 
the pilgrims to the tomb except on rare occasions when the pilgrims 
were influential. The leaders of the saintly lineages were usually care¬ 
ful to entertain important foqra who visited their ancestor’s tomb. 
Both the mizwars and other saints’ descendants made an exaggerated 


PILGRIMS 

(POOR AND AILING) 

* i 



GENERAL POPULATION 

FIGURE 6. 





The muqaddim of a Hamadsha team. 







The mausoleum of Sidi ‘Ali ben Hamdush in Beni Rachid. The tomb is located under 
the cupola; the wing, to the right, contains cubicles for pilgrims wishing to spend 
the night. 



Pilgrims entering the mausoleum of Sidi Ahmed Dghughi in Beni Ouarad. The tents 
belong to pilgrims. 







I 

I 


! 




The musem at Beni Rachid in 1968. The 
two palm trees are those at which the miz- 
wars of the saintly lineages meet. The 
mizwar of Sidi *Ali entertains guests in 
the black tent. The mosque of Beni Jcn- 
nad, a neighboring town, is in the back¬ 
ground. 


Circles of dancing pilgrims following the 
mizwar of Sidi Ahmed, hidden by flags, 
as he makes his way to the tomb of Sidi 
Mejma‘. Musem, 1968. 











Pilgrims at the bread ovens at Beni Rachid during the musem of 1968. Bread is 
symbolic of baraka and is said to contain it. 



Pilgrims at the ‘Ayn Kabir in Beni Rachid during the musem of 1968. The stairs 
lead down to the bathhouse, where women bathe in the holy waters . 



The hadra at Beni Ouarad. The man in the center of the circle has heard his rih and is 
about to enter jidba. Note the clay drums ( gwals ). 


The Hamadsha hadra before the mausoleum of Sheikh al-Kamal in Meknes. Note the 
long ghita in the upper lefthand corner of the crowd. 







Dghughiyyin of the madina of Mcknes recite the dikr. 


The zawiya of the Dghughiyyin in Meknes. Note the fig tree, to which bits of cloth have 
been tied as promises of a sacrifice (*ar) to either Sidi Ahmed or ‘Aisha Qandisha. The 
floor of the zawiya is covered with tombstones. 







A performance of the hadra by a bidonville 
team. The man with his head exposed is 
slashing it with a barely visible knife in his 
right hand. Note the varying responses of 
the audience. Note too the snare drum. 


A blow-up of the head-slasher. 










The “line” at the performance on the previous page. The men are in hal. The woman in 
the striped jallaba has joined the line “to come down.” 



Country Hamadsha performing the hadra in Meknes. The woman with flying curls is in 
jidba. The boy in the foreground hits his head with a shoe in imitation of a head-slasher. 



THE CIRCLE OF EXCHANGE 


123 


display of their hospitality during the musem, but although they did 
distribute bread and couscous and fed the important foqra, they actu¬ 
ally entertained only important visitors. 

The exchanges may be represented in Figure 6. Once again the thin 
lines indicate the direction of flow of wealth, and the thick ones that 
of baraka. Double bars represent an interruption in the flow. As can 
readily be seen, the redistributive system is interrupted by the mizwar 
and the team members, resulting in their accumulation of wealth. Al¬ 
though there may be a corresponding playing down of the team mem¬ 
ber’s baraka, baraka from the saint and his descendants is still dis¬ 
tributed. 

Indeed, it is precisely the conceptualization of the saint—and, by 
extension, the mizwar and other children of the saint—as a continual 
and inexhaustible source of baraka which enables the mizwar and the 
most influential of the saint’s descendants to exploit the system for 
their own ends. The social and concomitant ritual organization sup¬ 
ports their institutional endowment of baraka. It provides the mizwar 
with sufficient role distance to prevent his own personality, his selfish 
motives, and his prosaic character from interfering with his status as 
mizwar and his image as a selfless individual, highly endowed with 
baraka, who lives up to the highest ethical standards of Islam. More¬ 
over, the mizwar reconfirms his status during the musem not only by 
riding through the crowd of pilgrims but also by taking advantage of 
a dramatic opportunity to demonstrate his hospitality. 

The extent to which the mizwar is able to admit to the fact that he 
does not live up to the standards ideally expected of him is impossible 
to ascertain. It was my impression, however, that both mizwars were 
unable to accept fully the contradiction between their actual behavior 
and the behavior ideally expected of them. It has to be remembered 
that they themselves are members of the culture, subject to its values 
and standardized fears, and hold their ancestor in awe. This, I think, 
accounts in part for their exaggerated show of hospitality during the 
musem—a display which functioned for them as well as for the other 
pilgrims as a reconfirmation of the fact that they were living up to 
the ideal standards of mizwar behavior. 

Structurally insulated and ideologically supported by the widespread 
faith in their ancestors’ baraka, the mizwars were able to maintain their 
economically advantageous position without losing, at any rate outside 
their descent group, the image of pious and charitable men endowed 
with great baraka of their own. Within their own lineages, however, 
their images were far from pure. The wulad siyyid were well aware 



124 


THE HAMADSHA 


of the mizwars’ duplicity, but, as interested parties—eligible, theoreti¬ 
cally, at least, for a share in the proceeds—they were not so anxious to 
cast aspersion on the mizwars’ image, and preferred to vie one with 
another for their favor. 

Outside the saintly descent group, there was, however, comparatively 
little acknowledgment of the mizwars’ failure to live up to the obliga¬ 
tions of their role. The few team members who knew the facts were 
able to accept them quite readily. In fact, this became a good excuse 
for them not to give to their mizwars. The foqra who knew were much 
more reticent in admitting it. Even allowing for the image of their 
organization they wanted to communicate, their fear of their mizwar, 
and other considerations, there was resistance on their part to the realis¬ 
tic acceptance of the situation. Perhaps, were they to have accepted 
fully the mizwars’ failure to distribute the proceeds according to cus¬ 
tom, their belief in the system would have been shattered by the dis¬ 
sonant situation thus created, and they would have been unable to 
justify pouring their earnings into the mizwars’ hands. They chose, 
instead, to accept the ideal rather than the real. 

Beyond the positive injunction to give to the saint because one has 
received baraka from him, there are, as we have seen, certain negative 
consequences associated with the failure to give to him: business losses, 
paralysis, disease, and general misfortune. Here, the foqra’s belief was 
also threatened by the bidonville teams, which did not give to the saint 
as they should have and suffered no such dire consequences. The teams 
made no secret of their refusal to give all they could to the saints, but 
they were not in the same structurally advantageous position of insula¬ 
tion as their mizwars. The foqra argued that the bidonville team mem¬ 
bers were not true Hamadsha and that their failure to give to the saint 
accounted for their abject poverty. A few insisted that the team mem¬ 
bers were insane ( hmqin ). This, again, is a typical reaction to a dis¬ 
sonant situation. 

Only partially successful in resolving it, the foqra expressed a great 
deal of resentment toward the team members and refused to identify 
with them. The team members, who preferred their own economically 
advantageous status, had no desire to join the foqra. The mizwars, cal¬ 
lously manipulating the team members for what they were worth, 
encouraged the foqra’s resentment and discouraged any liaison between 
the two groups. The situation was, in part, responsible for the con¬ 
tinuing split between the foqra of the madina and the team members of 
the bidonville. 

The muhibbin, on the whole, were less interested in the intrigues 



THE CIRCLE OF EXCHANGE 


125 


within the Hamadsha organization than in the efficacy of the perform¬ 
ances. (One could also add, the dramatic quality.) They would often 
criticize both the team members and the mizwars, but in critical situa¬ 
tions they would always accept the baraka of the mizwar and the hadra 
of the team members. Often, they believed that 'A'isha Qandisha or an¬ 
other jinn had caused their “initiating” illness because they had scoffed 
at the Hamadsha. 

Although the muhibbin were not as partisan as the adepts or the 
team members, they tended to affiliate themselves with either the lodges 
of the madina or the teams of the shantytowns. The madina devotees, 
whose families had lived in the city for a few generations and who did 
not need to perform the hadra very often, would have nothing to do 
with the shantytown teams. They found their performances crass, 
spectacular, and not particularly pious. They accepted the prejudices 
of the foqra. The bidonville devotees, as well as the newcomers to the 
madina, preferred the teams. They explained that the foqra did not 
perform often enough, that they did not know the songs ( rihs) of the 
country, and that they chanted too long and danced too short a time. 
Some added that the foqra were not particularly friendly. By this, they 
meant that while the foqra treated them hospitably during a ceremony, 
the same foqra tended to ignore them on other occasions. The bidon¬ 
ville dwellers and the newcomers to the city seem to have expected 
more from their affiliation with the Hamadsha than the townspeople 
did. 

The Organization of Devotees 

There is also a significant difference between the organization of the 
madina devotees and that of the bidonville devotees. Team members in 
the bidonvilles play a much more active role in initiating ceremonies 
either directly or indirectly (through warnings to muhibbin said in a 
joking manner), and in determining the composition of ceremonies, 
than do the foqra. In the bidonvilles, clusters, determined by the fre¬ 
quency of invitations to one another’s ceremonies, tend to be focused 
on a member of one of the teams—usually the muqaddim or ghita 
player. He is, then, the core, or essential member of the cluster (cf. 
Barnes 1968:118). It is around him that much of the cluster’s activities 
takes place. In fact, each muqaddim or ghita player is the essential 
member of a number of clusters which are differentiated primarily by 
location. In the madina, clusters tend not to focus on an adept of a 
zawiya or any other single individual. The foqra are invited to perform, 



126 


THE HAMADSHA 


but they do not usually sponsor or invite other devotees to ceremonies 
in which they participate. The difference may be represented schemati¬ 
cally, as shown in Figure 7. The center of each star is either the team 
or zawiya or a member of either organization. The points of the star 
represent devotees who sponsor ceremonies from time to time. The 
arrows indicate the direction of invitations (cf. Barnes 1968:113). 

This difference in organization can be explained in part in terms of 
the greater economic stake the team members have in their perform¬ 
ances. Since they are economically dependent on the hadra, it is to 
their advantage to maximize their incomes by sponsoring ceremonies 
and by making sure that as many devotees as possible attend each of 
their performances. 

A comparison of the two muhibbin groups suggests another, latent 
function for this organizational difference. The devotees of the madina 
are, in fact, integrated into urban life. They have usually lived in the 
city all their lives, as did their parents and often their grandparents. 
They have professions, family, friends, neighbors of long standing, and 
other more or less permanent relationships. These relations tend to be 
specific, for specific ends. They are, as Max Gluckman puts it, single- 
stranded. The people of the bidonville are not integrated into urban 
life—they are in fact quite outside it—and they are not yet part of an 
established shantytown community. The shantytowns are too recent a 
phenomenon to provide their inhabitants with complete social net¬ 
works. The average bidonville dweller and his family came from the 
countryside without a profession that was in demand and without 
family, friends, or other contacts in the city. He often settled into a 
neighborhood in which he had no contacts and looked for work wher- 


• • 



• • 

MADINA SHANTYTOWN 

FIGURE 7. 



THE CIRCLE OF EXCHANGE 


127 


ever he could find it. The single thing he had in common with other 
people in his bidonville was devotion to the Hamadsha. It is not unusual 
to find that many of the shantytown devotees were actually inactive 
as Hamadsha before coming to the city. Although their sudden recom¬ 
mitment can be explained away on psychological grounds—increased 
tensions and pressures, feelings of isolation, loneliness, and unworthi¬ 
ness—the facts suggest that the status of devotee provided the new¬ 
comer with a means of enlarging his social network. 

While the devotees of the madina depend upon their own contacts 
for inviting and being invited to the ceremonies, those of the shanty¬ 
towns depend less on their own personal contacts than on the contacts 
of the team members for their ceremonial attendance. The team mem¬ 
bers, who are professional Hamadsha, have a wide range of these con¬ 
tacts and serve, if you will, as mixers. They provide the devotee with a 
large number of contacts with whom he will develop reciprocal rela¬ 
tions—mutual invitations and the like—not only within the Hamadsha 
frame of reference but also within the broader social setting. Again and 
again, analysis of the life histories of the bidonville Hamadsha reveals 
that their most significant contacts outside family and kin within the 
city of Meknes were other Hamadsha. The links between the devotees 
of the bidonville are not as single-stranded as those of the madina. 
They tend to be multiplex. They satisfy many more needs of the new¬ 
comers, who come from the country where multiplex links are pre¬ 
dominant, than do the single-stranded links of the madina dweller. 

The reluctance of the bidonville devotees to affiliate themselves with 
the madina zawiyas is, moreover, a reflection of the breach between 
city and country—city and bidonville—discussed in the last chapter. 
Certainly there are ritual differences between the two muhibbin groups, 
but the rituals themselves are not rigid and could easily be modified. 
Rather, the concern for ritual differences seems to symbolize the dif¬ 
ference between madina and bidonville inhabitants which the bidon¬ 
ville dweller senses. He prefers the bidonville teams because their dev¬ 
otees are very much in the same position as himself. Often, too, the 
teams’ performances are near his own home and provide him with 
regular occasions to perform the hadra. Even newcomers to Meknes 
who have settled in the madina and attend zawiya performances prefer 
performances in the bidonvilles and go to them whenever they can. 
The foqra, who consider themselves city people, have never been 
anxious to develop a large following of bidonville muhibbin, despite 
the obvious increase in income these devotees would bring. Whether 
this reflects a fear of economic takeover or the desire not to be identi¬ 
fied with the newcomers cannot be determined. The foqra themselves. 



128 


THE HAMADSHA 


it has to be remembered, are comparative newcomers to the city too. 
One point is obvious—the majority of foqra are elderly, and they can¬ 
not, physically, perform the hadra as often as a large following would 
require. With such a large following, the foqra would see their position 
gradually usurped by the bidonville teams. The team members would 
not only pose a threat to the foqra’s status within the organization; but, 
as we have seen, they would affect the very basis of their belief and 
practice. 



Part Three 


The following chapters are concerned with Hamadsha therapy. I have 
chosen to present my material within a framework that will be familiar 
to readers of Western medical and psychiatric literature, because such 
a framework provides both a means of describing the Hamadsha system 
of therapy with minimum distortion and a suitable vehicle for compari¬ 
son with both Western medical and psychiatric theories and those of 
other societies. It has to be remembered, however, that this presentation 
is not the Hamadsha’s, nor would the Hamadsha feel altogether com¬ 
fortable with it. As participants in a system which they have found to 
be more or less successful therapeutically, they have not felt the need 
to organize their beliefs and practices into a consistent whole. Were 
they to attempt it, they undoubtedly would not delimit their “thera¬ 
peutic system” in the same way I have done. Their axis of orientation 
would be quite different. It would probably emphasize many more ele¬ 
ments of what the Westerner would call religion or folk theology. 

The Hamadsha system of therapy is not an isolated, secular system 
of beliefs and practices; it is embedded in and indivisible from the entire 
culture of the Moroccan Arab. The lines between secular and sacred 
have not been drawn, and Moroccan culture has not experienced the 
fragmentation resulting from specialization and overspecialization that 
Western culture has undergone. Throughout my presentation I have 
had to bring in data from other areas of Moroccan cultural life. The 
Hamadsha’s therapy, like all therapies, is a legitimating apparatus—“the 
application of conceptual machinery to ensure that actual and potential 
deviants stay within the institutionalized definitions of reality” (Berger 
and Luckmann 1967:112-113)—and their psychology, like all psychol¬ 
ogies, presupposes a cosmology (Berger and Luckmann 1967:175). 



8 


The 

Theory of 
Therapy 


Hamadsha therapy is but one of many therapies within the Moroccan 
world, a point which has to be emphasized. Although in the Western 
world there are a number of quasi-legitimate therapies (fringe medi¬ 
cine) such as chiropractic and osteopathy, there is only one standard 
system of therapy—medical therapy—which has received the “charter” 
of the society. When a Westerner is sick, he will almost invariably have 
recourse to some product or some practioner within the medical field. 
The doctor of medicine, particularly the specialist, is the final authority. 

In Morocco, there is no single, socially chartered therapeutic system 
with final authority. There are the fuqaha, or Koranic teachers, who 
specialize in writing amulets and talismans; the herbalists, who may be 
either Arab, Berber, or Jewish; the specialist in traditional Arab medi¬ 
cine; the barber, who, among other things, lets blood; the aguza , an 
old woman who is familiar with many herbal and magical brews and 
practices and who may be a midwife; her male counterpart; the exor¬ 
cist, who may be a Koranic teacher or an adept of a brotherhood; the 
brotherhoods themselves—Gnawiyya, Tsawiyya, Rahaliyya, Hamdu- 
shiyya, Jilaliyya, and the like; and Western medical practitioners. Even 
the Western medical practitioners are not popularly organized into a 
hierarchy of authority, and the Moroccan’s decision to go to one rather 
than another is as significant as his decision to go to an herbalist rather 
than a Koranic teacher. There is, then, the choice of going to a French 
or Spanish doctor, to a Moroccan doctor who has studied Western 
medicine, to a pharmacist, to a male nurse who conducts a medical 
practice on the sly, to a European missionary, or to a local infirmarv 



134 


THE HA MADS HA 


or hospital. The Moroccan may also visit any number of saints’ tombs 
or sacred grottos, springs, pits, trees, or baths. 

Although symptomatology plays a definite role in the choice of ther¬ 
apy, it is not the only consideration, especially since several different 
types of therapy are usually believed to be efficacious for the same 
complaint. Other factors are of equal, if not greater, importance. These 
include notions of what has caused the illness, the availability of the 
specialist, the cost of therapy, 1 the nature of the relations the patient 
or his family has with various practitioners, the past experiences of the 
patient and his family with various cures, the urgings of friends and 
neighbors, the advice of diviners and seers, the current popularity of a 
particular practitioner or type of therapy, the strength of belief in a 
particular saint, shrine, or brotherhood, and the seriousness and urgency 
of the complaint. The wide range of options leaves considerable room 
for unconscious as well as conscious factors in the choice of cure. 

The Moroccans do not make the same distinction between physical 
and mental illness that the Westerner does. Just as they have many dif¬ 
ferent forms of therapy, so they have many theories of causation. Their 
“pathologies” arc often self-contradictory. They appear to be an amal¬ 
gamation of various folk beliefs and popularizations of traditional Arab 
and modem Western medicine. Since theories of causation vary from 
tribe to tribe, city to city, and family to family, it is impossible to form 
a systematic Moroccan pathology. Still, although a supernatural ele¬ 
ment will enter either directly or indirectly into the Moroccan’s ex¬ 
planation of any disease, it is possible to divide Moroccan theories of 
causation roughly into two categories: naturalistic and pretematuralistic 
explanations. 

The naturalistic explanations are usually mechanistic. Syphilis is said 
to be caused by menstrual fluid. Gonorrhea is said to be caused by 
semen backing up into the kidneys, if a man is not careful to urinate 
after making love. Tuberculosis is caused by, among other things, not 
wearing a shirt in the heat, especially when riding a bicycle. Certain 
heart diseases are said to be caused by eating salt. Poisoning by one’s 
enemies is a frequent explanation for wasting diseases. 

Preternatural explanations can be divided into those involving the 
jnun and those that do not. The latter include magical poisonings, magi- 

1 Pilgrimages to saints’ tombs and other traditional therapeutic practices are 
often chosen over Western therapies because of cost. This is not just true of 
Morocco but of other countries in North Africa and the Middle East (cf. Howell 
1970). 



THE THEORY OF THERAPY 


135 


cal curses, witchcraft, and the evil eye. The former have to do with 
being struck or possessed by a jinn. 

It is for illnesses caused by the jnun that the Hamadsha are called, 
and it is with these that we will be primarily concerned. It should be 
remembered, though, that the cults to Sidi c Ali and Sidi Ahmed have a 
broader range of efficacy. Although visits to the tombs of the Hamad¬ 
sha saints are considered particularly helpful for illnesses caused by the 
jnun, these visits are also held to be beneficial in curing illnesses for 
which the jnun are not responsible. The pilgrim to the tombs obtains 
the baraka of the saints and somehow always benefits thereby, even if 
he is not entirely cured of his complaint. 

I wish to emphasize, too, that illnesses caused by the jnun are not the 
sole province of the Hamadsha. Throughout Morocco, the fuqaha are 
considered to have special control over the jnun. They provide men, 
women, and children with talismans to protect them from the jnun, 
and are often called in to exorcize devils who have taken possession of 
a person. The members of the other fraternities—most notably the 
Gnawa, from whom the Hamadsha have borrowed heavily—also spe¬ 
cialize in the cure of the jinn-struck and the jinn-possessed. The deter¬ 
mination of which particular cure to undertake varies from case to case. 
Often, the patient will try one cure after another until he happens upon 
the one that “satisfies the jinn”—that is, the one that offers him relief. 

The Theory of Pathology: The Jnun 

The term jnun, the Moroccan Arabic plural of jinn, which is itself 
a plural in Classical and many dialects of Arabic, has both a general and 
a specific meaning. Generally, it refers to a wide variety of poorly dif¬ 
ferentiated “spirits” which are believed to inhabit the universe, but 
which are distinct from the ordinary beings of everyday life. These 
spirits are of a different ontological status from the objects of the phe¬ 
nomenal world. The exact nature of this difference is not clear. Their 
existence is attested to in the Koran, where it is written that God 
created man "of clay like the potter’s” and the jinn “of smokeless fire” 
(LV, 13-14). The Westerner would classify them as supernatural or 
elemental spirits. 

There are many terms for jnun in Moroccan Arabic. Some of them 
refer to specific classes of jnun—* afarit, ghwal, and shayatin —while 
others, as Westermarck has noted, are the result of fear of mentioning 
them by their proper name. “To pronounce their name would be to 



136 


THE HAMADSHA 


summon them” (Wcstermarck 1926(1):263). I have heard them re¬ 
ferred to as mink (the plural of malik, king), muselmin (Muslims), 
muvoalin l-ard (the masters of the earth), mwwalin l-makan (the mas¬ 
ters of the place), rjal al-blad (the men of the country, a term that is 
used most often for saints), and mu’minin. The term l-aryah, which is 
the plural of rib (wind, or, in the case of the Hamadsha, the particular 
trance-inducing musical phrase of a person), is very commonly used. 
Westermarck (1926(1):263) found that its usage was restricted to 
disease-causing jnun. Most often, however, references to the jnun are 
oblique: “those people there,” “those people who arc below the 
ground,” “those people who are below the river,” “those who shun 
salt,” or “those who are hidden or invisible.” French-speaking Moroc¬ 
cans refer to the jnun as les diables or les invisibles. 

Not only are the lines of distinction between the various classes of 
jnun unclear, but the jnun themselves are often confused with other 
spiritual beings such as angels ( maldika) and Satan (Shitan, iblis ), as 
well as with the saints themselves. Several of my informants confused 
the jnun with the two angels which are believed to live on a person’s 
right and left shoulders and to count, respectively, their good and bad 
activities. Saints, angels, and jnun are often lumped together, as when 
the expression rjal al-blad, which commonly refers to saints, is used for 
the jnun, too, especially for those which I call named-jnun. Wester¬ 
marck (1926(1):389) has noted this confusion between saints and jnun. 
He writes: 

Saints rule over jnun and have Muhammadan jnun as their assistants: and 
Mulai ‘Abdelqader [Jilani] is the sultan of all the saints and jnun. The term 
rijal allah comprises both saints and Muhammadan jnun, and in the jenn 
saints the borderline between jnun and saints is wellnigh obliterated. 

Westermarck goes on to list animals such as frogs and snakes which 
may be considered either saints or inhabited by jnun, and the shrines 
associated with them. 

Although the term jnun may be used for the ‘ afarit, ghvoal, and 
shay atm, most Moroccans readily acknowledge a difference, properly 
speaking, between these creatures and the jnun. The shayatin are asso¬ 
ciated with Shitan, or Satan, and have been called his children. Like 
Shitan, their status in the spirit world is not certain. The Koran refers 
to Shitan as both a jinn (XVIII, 48) and an angel (II, 32). Most 
Moroccans hold the shayatin responsible for mischievous thoughts and 
deeds, especially those deeds that appear to be more or less spontaneous 



THE THEORY OF THERAPY 


137 


and are against the better judgment of the man who perpetrates them. 
The shayatin are often blamed for sexual misconduct. 2 A man will speak 
of his amorous adventures with reference to the shayatin, or he will 
refer to women as shayatin when he reflects on their supposedly inex¬ 
haustible appetites and their inability to counter their libidinous desires. 
The transgressions for which Shitan or the shayatin are held account¬ 
able do not inspire strong feelings of remorse or guilt; the shayatin are 
not responsible for diseases of any sort. 

The jnun, in the restricted sense of the term, also have to be dis¬ 
tinguished from the huge, cannibalistic spirits known as c afarit, and the 
ghwal, or ogres. Although the Koran takes notice of the jinn and 
Shitan, it refers neither to the 'afarit or the ghwal. 3 Both the 'afarit and 
the ghwal appear to have a more substantial existence than the jnun, 
whose bodies are said to be made of vapor or flame. The ‘afarit are, 
perhaps, more powerful, clever, and evil than the “run-of-the-mill” 
jnun. Westermarck refers to them as the “aristocracy of the jnun.” It 
is my impression that they occupy space more often, at least, than the 
ordinary jnun, and are capable of being perceived by man’s ordinary 
perceptual faculties. They seem to differ in scale and capacity from 
both man and the other jnun. They may be giants with tremendous, 
often anthropophagous appetites; they may be capable of traveling 
great distances with extraordinary rapidity; or they may be endowed 
with great strength. They are always malignant and terrifying. They 
have the ability to be in more than one place at a time, to fly, and to 
render themselves invisible. They may possess a person, thereby ren¬ 
dering him a maniac with extraordinary strength. 

The ghwal, or ogres, may be either male ( ghul ) or female ( ghula ). 
Belief in them is widespread in the Arab world. The ancient Arabs be¬ 
lieved that they lived in the desert and, assuming the shape of women, 
led travelers astray. They appeared, and still appear in Morocco, more 
often as women than as men. Usually they are giant women with long 
pendant breasts and the feet of a goat. They are sometimes covered 
with hair. Although they are usually ugly, they have the ability to 
change their shape and become beautiful, enticing women, who lure 
their victims into deserted places and eat them up. Westermarck 
(1926(1): 397) has written of them: 


*The association of the shayatin with sexual activity is exemplified by the idio¬ 
matic expressions for sexual intercourse during dreams: darab sh-Shitan and 
ja sh-Shitan . 

* Line 39 of Sura 27 refers to an “ £ ifrit of the jinn,” but this is probably a figure 
of speech which means a very powerful jinn. 




138 


THE HAAIADSHA 


The glnval are said to live in the Sudan or Sahara or in a thick wood from 
which they come out in the shapes of animals or who knows where? They 
have black faces and eyes like flaming fire and are fond of human flesh. 
Some people maintain that ghival are not jnun but a species by themselves 
or a kind of men or wild animals, whereas others, who seem to be particu¬ 
larly well-informed, are of the opinion that they are of the jerm kind. 

The female ghwal possess many of 'Aisha Qandisha’s features, and I 
have heard her called a ghula. Usually, however, she is referred to as a 
jinniyya or, especially by her most constant devotees, as simply Lalla 
c A'isha, a creature different from all other creatures. 

The jnun—again in the more restricted sense of the word—are a race 
of spiritual beings, created before man and said to be composed of 
vapor or flame. They arc not normally extended over space, but they 
do exist over time. They are intelligent creatures, resembling human 
beings, but they have no bodies and are usually imperceptible to man’s 
ordinary senses. They are, however, capable of rendering themselves 
visible, and often take on various animal forms: snakes, frogs, wasps, 
cats, camels, donkeys, and jackals. They sometimes appear as female 
temptresses or male seducers. The jnun are capable of marrying hu¬ 
mans, and their legal status has been discussed in all respects by Muslim 
theologians and canonists. They are also capable of appearing in more 
than one place at a time. There is, in Morocco, mixed opinion as to 
whether or not the jnun are mortal. Most Moroccans believe that Mus¬ 
lim jnun are capable of salvation. 

The jnun are said to live underground, although some of them are 
believed to inhabit the air between the earth and the sky. They are said 
to like water and to gravitate to damp or wet places: marshes and bogs, 
streams and rivers, springs and wells, and the sea. They are frequently 
found near fountains, baths, toilets, and drains. Often, too, they live in 
grottos and caves or around unusual trees and rocks. All the areas where 
jnun are said to live are treated circumspectly. A Moroccan who ap¬ 
proaches such a place will usually say bis?mflah r-rahman r-rahim, in 
the name of God the merciful and compassionate, or simply bismillah , 
to rid the area of their presence. He will also sprinkle salt, to which the 
jnun have a special aversion, around toilets and drains. The jnun have 
an equally intense dislike of benzoin ( jawi ) and other incenses, 4 and of 
iron and steel, silver, and tar. 

4 Nevertheless, under certain circumstances, as we shall see, the jnun are at¬ 
tracted by or pleased with jawi and other incense which normally drive them off 
(Paques 1964:559-563). 



THE THEORY OF THERAPY 


139 


The jnun do not usually have characters as well-developed as, say, 
the Zar of Ethiopia (Leiris 1958) or the Vodun spirits of Haiti (Me- 
traux 1959). They cannot be equated with natural phenomena such as 
lightning and thunder or stars and comets, although certain of the 
c afarit are held responsible for these phenomena. Some of the jnun are 
named, and their names are well-known to all Moroccans. Westermarck 
(1926(I):49) has called these jnun “jenn saints.” Others—the majority 
—are named, but their names are the secret of a few privileged individ¬ 
uals who have access to their world. There is much power in the name, 
and the discovery of the name of a jinn possessing a person is one of 
the first steps in exorcising it. 

There are male and female jnun; Muslim (Arab and Berber), Jewish, 
Christian, and pagan jnun. The pagan jnun are especially dangerous. 
All jnun are capable of marrying among themselves and bearing chil¬ 
dren. It is claimed, in fact, that the jnun are organized into a world 
which mirrors the Moroccan world. There is a king of the jnun, Sidi 
Shamharush, who is venerated as a saint in some parts of the country, 
around Marrakech, 5 as well as a court of ministers, pashas, and other 
traditional civil servants. A Moroccan does not push the parallel very 
far, however, and members of the jnun world do not normally have 
relations with their counterpart in the phenomenal world. 

Sidi Shamharush is said to have a complicated communications net¬ 
work that is likened to a river with a thousand tributaries, each of 
which has a thousand sub tributaries. There is a jinn responsible for each 
of these rivers, who runs errands and carries messages for the king of 
the jnun to the farthest reaches of the world. The jinn of a river is 
named after the river; the names of these jnun are known only to a 
few, the Koranic teachers and magicians, and their knowledge gives 
them a certain control over the jnun. These magician-scholars are said 
to use their power to find buried treasures—a very frequent theme in 
Moroccan folklore—and to work magic and exorcise demons. 

Often the jnun make their appearance in dreams, but it is possible by 
various practices involving blood, to which the jnun have a special 
affinity, to obtain a view of them. The jnun are frequently said to carry 
men and women off to their world and reveal its workings. It is not 
unusual to hear of a man suddenly disappearing from his village and 
then returning many years later to tell of his adventures among the 
jnun. Many of the magician-scholars are said to leave the phenomenal 


8 He is believed to be buried at the foot of the Jebel Toukhal. 



HO 


THE HAMADSHA 


world for varying periods of time and to descend into the rivers, where 
many of these spirits gather. They return with considerable power and 
are treated with care. 

The jnun are not necessarily evil or harmful. They are whimsical 
and arbitrary, capricious and revengeful, quick-tempered and despotic 
—and therefore always potentially dangerous. If they are wounded or 
insulted, they are quick to retaliate by striking their adversary or taking 
possession of him. The man, or woman, who has injured a jinn is usu¬ 
ally quite ignorant of his misdeed. He may have brushed against it 
accidentally, trampled it inadvertently—particularly when the jinn has 
taken on the form of an animal such as a snake or a frog—or scalded it 
with boiling water which he, or more commonly she, has carelessly 
thrown out. It is often said that where a jinn is injured, there will his 
injurer suffer. 

Men and women who are angry or frightened are particularly liable 
to attack by the jnun; so are people in liminal periods associated with 
change in social status. Hence pregnant women, newborn children, 
boys about to be circumcised or just circumcised, couples about to be 
married, and the dying are particularly vulnerable. There are periods 
when the jnun are especially active—after mid-afternoon prayers, for 
example. There are also periods when the jnun are especially inactive— 
during Ramadan, the month of fasting, when the jnun are said to be 
imprisoned. 

The jnun are always treated with the greatest respect, and they are 
generally feared. Moroccans take all kinds of precautions to keep them 
at bay. They may burn candles in the dark, sprinkle salt in areas to 
which the jnun are known to gravitate, utter apotropaic phrases, when 
crossing streams or thresholds, entering caves, automobiles, and un¬ 
familiar places, or simply feeling the presence of a jinn; and they usu¬ 
ally wear phylacteries containing words from the Koran as well as 
charms to protect themselves (cf. Westermarck 1926(1):302-324). The 
jnun’s constant interference with man’s daily life is accepted fatalisti¬ 
cally, however, as one of the givens of wordly existence. 


Named-Jnun 

There are, as I have said, certain jnun whose names are common 
knowledge and whom I have chosen to call “named-jnun.” They are 
not linguistically distinguished in Moroccan Arabic, as far as I know, 



THE THEORY OF THERAPY 


141 


but are always referred to by their proper names. 8 Westermarck has 
classified many of them as individual spirits. Because the named-jnun, 
especially c Ai'sha Qandisha, play an important role in Hamadsha ther¬ 
apy, it is necessary to consider them and their relationship with human 
beings in detail. 

With the exception of the belief in ‘Ai'sha Qandisha, belief in the 
named-jnun in Morocco is associated primarily with the Gnawa. The 
Gnawa, like the other so-called black brotherhoods of the Maghreb, 
have an elaborate hierarchy of demons, which are associated with spe¬ 
cial colors, incense, dance steps, and musical phrases. 7 There is some 
character differentiation among them (Dermcnghem 1954; Paques 
1964). The named-jnun are also associated with the Jilaliyya—the more 
“popular” and spectacular of the two brotherhoods that claim to be 
followers of al-Jilani. (The Jilala, however, are often confused with 
the Gnawa in Morocco.) Brunei (1926:145 et seq.) mentions the be¬ 
lief in named-jnun among the Isawa. The Hamadsha themselves, as 
well as most other Moroccans of their background, recognize the 
Gnawa origin of many of their named-jnun. 

Westermarck has argued (1926(1): 379 et seq.; also see Paques 1964) 
that the belief in these named-jnun was brought north from sub- 
Saharan Africa by black African slaves. The word “Gnawa” is said to 
be derived from Guinea. Tremeame (1914) has traced similar Tunisian 
beliefs to the Hausa; Andrews (1903) found them among the Songhrai 
immigrants to Algiers. I myself was told both by members of the 


* In many of the religions of Northwest Africa, south of the Sahara, a distinction 
is made linguistically between the named and unnamed spirits (Greenberg 1946; 
Ortigues 1966). 

7 Viviana Paques, in her extraordinary monograph, UArbre Cosmique (1964: 
568), examines the symbolic significance of these associated elements in terms of 
an esoteric “African” cosmology. She considers the dances symbolic, recalling the 
first mythic and cosmic events which brought about the formation of the world 
in which we live. Bastide (1958:175) has argued similarly for the condovible of 
Bahia. “La transe religieuse est reglee par les modeles mythiques; elle n’est qu’une 
repetition dcs mythes. La danse devient un ‘opera fabuleux; l’expression celebre 
de Rimbaud ‘colie’ exactement au phenomene.” 

Among the Hamadsha, and among other Moroccans of their milieu, I have 
found no such symbolic associations—and no knowledge even of the existence of 
such estoteric beliefs of which Paques writes. The Hamadsha consider their dances 
as work for the saints they worship and as an appeasement of the jnun. They 
recognize of course that they fall into trance ana mutilate themselves as Sidi 
Ahmed (and Sidi f Ali) did. This is inevitable. It is the will of Allah, and perhaps 
the will of the I^amadsha saints and the jnun, even. They sec it as a mark of 
devotion, of worship, even of obligation, but not as the symbolically significant 
reenactment of the practices of the saints they worship. 



142 


THE HAMADSHA 


Gnawa order and by nonmembers living in the Valley of the Sous (at 
Sidi Ahmed ou Alousa and at Illergh) that the belief in these named- 
jnun came from the Sudan. Legends like that of c Ai‘sha Qandisha sug¬ 
gest a Sudanese origin. Parallels between the religions just south of the 
Sahara—the religions of the Hausa, the Songhai, and the Zar worship¬ 
pers in Ethiopia—and the belief in named-jnun among the Gnawa, 
Jilala, Hamadsha, and c Isawa are striking. 

The extent to which the belief in named-jnun and the development 
of specific relationships to them is restricted to the followers of these 
brotherhoods cannot be determined. Michaux-Bellaire (1913:294) 
found belief in them current among the women of the Gharb in 1913. 
In fact, he suggested that this belief was restricted to women; but it 
exists today among both men and women affiliated with the Hamadsha 
and other similar brotherhoods. Still, it is claimed that more women 
than men have intimate relations with the named-jnun, especially with 
the female ones. The section of the hadra devoted to these demons is 
said to be for women. One of my best informants suggested that belief 
in them is restricted to Arabs of the northern countryside: Gharbawa, 
Beni Hsen, Shrarda, Hajawa, and Zrahna. Although there certainly is 
such a belief among these people, I am not certain that the belief is 
restricted to them. 

The named-jnun have all the attributes of ordinary jnun. Their 
character traits and personalities are more elaborate, but they are not 
as well worked out as those of the Vodun and Zar spirits. Some of 
them, like c A'isha Qandisha, have legends associated with them. Having 
names, they all have sexual identities. The majority are female, but 
there are a few males. Some of the males are married to the females 
and may even have children. The named-jnun often appear as incubi 
and succubi. (Often the unnamed do, too.) Unlike the ordinary jnun, 
they establish definite relations with men and women and make certain 
demands on them. 

The relationship between a man or a woman and his or her named 
jinn or jinniyya is complicated. Such a man is said to be the follower 
(tab?) of the jinn and to lean on ( muttakil) or rest against {muivali) 
the jinn. Msannad may also be used to describe the relationship be¬ 
tween a man and his jinn, although it is usually restricted to his rela¬ 
tionship with a saint. A man can decide—but does not necessarily de¬ 
cide—to become msannad to a saint; he cannot decide to become a 
follower of a jinn. It is the jinn, as we shall see, who always decides 
this. Once the relationship has been established, the jinn makes certain 



THE THEORY OF THERAPY 


143 


demands on his follower. These usually involve wearing a certain color, 
burning a special incense, and performing the hadra to a special musical 
phrase ( rih ). The jinn may also require special food and sexual taboos, 
visits to shrines, periodic sacrifices, or the sponsoring of ceremonies, 
especially those given by the Hamadsha, the Gnawa, and the Jilala. 

* A is ha Qandisha 8 

Of all the named jnun, ‘Aisha Qandisha is the most important for 
the Hamadsha. They consider themselves to be her special devotees. 
Often they refer to her as ‘“Ai'sha Sudaniyya,” ‘“Aisha Gnawiyya,” 
“Lalla ‘Aisha” (Lady ‘Aisha), or the great female muqaddim. I have 
also heard them call her “our mother.” Westermarck (1926(1): 395- 
396) identifies ‘Aisha Qandisha with Astarte, the goddess of love of the 
ancient Eastern Mediterranean. He suggests that “Qandisha” is related 
to “Qedesha,” the name for the temple harlot in the “Canaanitish cults,” 
and that belief in her was brought to Morocco by the early Phoenician 
invaders. He further suggests that Hammu Qiyu, ‘Aisha’s jinn-husband, 
can be identified with the Carthaginian god Hamam. While the belief 
in ‘Aisha Qandisha and the Hamadsha practices do show parallels with 
ancient Mediterranean beliefs and practices, there are equally strong 
parallels between the belief in ‘Aisha Qandisha and current beliefs in 
other female demons and ogresses throughout the Mediterranean basin 
and in sub-Saharan Africa (Frazer 1961). 

Although ‘Aisha Qandisha is generally referred to as a “jinniyya,” 
her exact status in the Moroccan demonology has never been deter¬ 
mined specifically. To the coastal Arabs she is a siren, and to those of 
the interior she is a sort of chthonian mother associated with earth, 
mud, water, and rivers. She is always libidinous and quick-tempered. 
She never laughs, and she is always ready to strangle, scratch, or whip 
anyone who insults her or does not obey her commands. It is said that 
she prefers the morning air for her walks. Like Kali-Parvati, she may 
appear to believers either as a beauty or as a hag with long pendant 
breasts. Usually, even in her beautiful manifestations, she has the feet 
of a camel, a donkey, or an ass. Sometimes she appears with the legs 
of a woman and the body of a goat with pendant breasts (Westermarck 
1926(1):392). Throughout northern Morocco there are places sacred 
to her. These are usually pits, grottos, springs, and fountains, as well 
as other spots where someone is said to have seen her or where some- 


8 See also pp. 43-45. 



144 


THE HAMADSHA 


thing uncanny has taken place. The Hamadsha, particularly Sidi 'Ali’s 
followers, say she resides principally in the grotto of Beni Rachid. 
However, she too is capable of appearing in many places at the same 
time. 

When ‘Aisha Qandisha presents herself to a man as a beautiful 
woman, a seductress, that man will have no defense against her power 
unless he immediately plunges a steel knife into the earth. He is then 
privileged to reject her entirely or to make a “marriage contract” with 
her that is to his own advantage. (This is infrequent, since Lalla 
'Aisha invariably hides her camel’s feet under a flowing caftan.) If a 
man has been unfortunate enough to sleep with her before discovering 
her identity, he becomes her slave forever and must follow her com¬ 
mands explicitly. Otherwise, she is sure to strangle him. 

A man is walking along a road, and suddenly his vision blurs. He thinks 
there is something wrong with his eyes, but in fact it is Lalla e AIsha Hasna- 
wiyya [one of e Aisha Qandisha’s manifestations]. He sees only her in front 
of him, and he looks on and sees only her. When he comes to an isolated 
crossing or path, she takes him by the hand. She asks him why he is follow¬ 
ing her and where he knows her from. The man tries to excuse himself and 
says that he thought she was a woman he knew. She says, “Fine. Welcome. 
Come with me. But we must go into a garden, and not into a house.” “I’ll 
even go beyond the garden,” the man answers. Lalla e Aisha agrees and tells 
the man to follow her. Because he follows her, he sees only clouds. He 
alone sees her. Suddenly they find themselves in a big garden with a lot of 
food, near a well-furnished house. Lalla 'Ai'sha takes the shape of a woman 
the man loves and desires. She has the same features as the beauty, but they 
are slightly exaggerated. Her bust is eighty centimeters. They make love. 
Afterward she asks him what he wants to do. The man says he is single and 
wants to marry. Lalla e Atsha says she wants to do the same. “We’ll make a 
vow [ ahd ] to God,” she says. She tells him not to tell anyone, not even 
his mother. If he does, she’ll have his throat. She tells him that she will 
sleep with other men. “I’ll help you with money,” she says. “You must not 
wear European clothes except if they are covered with a jellaba.” Then she 
asks him if he knows her. She tells him who she is. “If you do not do what 
I say,” she says, “then you’ll not have me or any other woman.” The man 
asks for three days to think it over. Lalla c Ai'sha gives him four. “If I don’t 
do this,” the man thinks to himself, “then I’ll have to dress in a darbala 
[a ragged and patched cloak]. I may even lose my life. I’ll marry her. 
Everyone knows me. I don’t want them to see me change.” 9 

8 This example of c AIsha’s ways was given me by one of my best informants, a 
bom storyteller, who spent hours entertaining friends with his talcs. He was not a 
Hamdushi but was very familiar with their ways. He was popularly thought to 



THE THEORY OF THERAPY 


145 


Often ‘Aisha will demand that the man wear old and dirty clothes re¬ 
gardless, and that he never cut his hair or fingernails. She will usually 
restrict his sexual life too. She may forbid him to make love to any 
woman but herself and his real wife, or to any woman who is not on 
her side, her follower. Such men are said to be married ( mja r umej) to 
‘Aisha Qandisha, and they are treated circumspectly. This special rela¬ 
tionship to ‘Aisha seems to bear no relation to membership in the Ha- 
madsha brotherhoods. In fact, only a few of ‘Aisha’s husbands are 
Hamadsha. 

‘Aisha Qandisha requires her followers—and not necessarily her hus¬ 
bands—always to wear red, black, or chartreuse green, or any combina¬ 
tion of these colors. Red and black are the more common. Her incense 
is black jawi (benzoin) or hasalban (?), which is used to placate her, 
especially when she takes possession of her follower. She prefers the 
music of the Hamadsha, and often requires her followers to perform 
the hadra. Her special rih has the following words: 

O 'Aisha! Rise and place yourself in the service of the cause of Allah and 
the Prophet. 

O Sire! Greetings to the Prophet. Welcome, O Lalla 'Aisha. 

The altar is prepared. O Lalla 'Aisha! O Gnawiyya! 

Welcome, O Daughter of the river. Allah! Allah! Lalla 'Aisha! 

She has come! She has come! She has come! Lalla 'Aisha! 

There is some variation in these words. 10 This rih is played at the peak 
of the Hamadsha hadra. The lights are turned out, and 'Aisha is said 
to rise from the earth under the dancers and dance with her devotees. 
People claim that whenever and wherever the Hamadsha perform 
their hadra, 'Aisha Qandisha is under them in the earth. She is par¬ 
ticularly fond of blood, and often requires her followers to make 
sacrifices ( dabihas ) to her. Black and red chickens are her most com¬ 
mon demands. She also makes certain that her followers slash their 
heads when they hear her special rih or another rih that pleases her. 
Still other of her followers are made to imitate pigs or camels when 
they perform the hadra. Such thcriomimctic behavior is even more 

common among the ‘Isawa (Brunei 1926:168 et seq.). 

Sometimes 'Aisha Qandisha fragments into a number of different 

be married to the she-demon, and he considered himself to be tormented by her. 
This tale, which is considerably embellished and not without humor and suspense, 
was told to me in front of several Hamadsha devotees who were very impressed 
by it. 

10 Wcstermarck (1926(I):393) records a verse emploved in the area of the Duk- 
kala tribe: “'Aisha Qandisha is sitting and smears herself with henna.” 



146 


THE HAMADSHA 


jinniyyas, each with a slightly different personality: Lalla c A'isha Su- 
daniyya, who is also called Lalla ‘Ai'sha Gnawiyya and is the “basic” 
‘Aisha Qandisha; Lalla ‘Ai'sha Dghughiyya; Lalla ‘Ai'sha Dghugha; 
and Lalla ‘Aisha Hasnawiyya. It is sometimes said that these jinniyyas 
are the daughters of ‘A'isha Qandisha, but most often they are simply 
named and their relationship to ‘A'isha Qandisha is left unquestioned. 
They all require their followers to wear black or red, but they appear 
to be placated by different incenses. Lalla ‘Aisha Dghughiyya likes red 
jawi; Lalla ‘A'isha Dghugha likes harmal (Paganum harmola ); and Lalla 
‘A'isha Hasnawiyya likes tar (qatran). I have been told that these 
“ladies” have different favorite times of day for taking walks. Lalla 
‘Aisha Dghughiyya likes the hours immedately following afternoon 
prayers; Lalla ‘A'isha Dghugha prefers the hours after evening prayers; 
Lalla ‘A'isha Hasnawiyya only the first few hours after evening pray¬ 
ers. Lalla ‘Aisha Dghughiyya supposedly comes from Sidi Ahmed’s 
village, and Lalla ‘Aisha Dghugha from the village of Sidi ‘Ali. (Lalla 
‘Aisha Dghugha is said to be particularly fond of houses.) Lalla ‘Aisha 
Hasnawiyya comes, as her name indicates, from the Beni Hsen. There 
is considerable variation in the extent to which these manifestations of 
‘A'isha Quandisha are developed. Many of the Hamadsha do not believe 
in them, and some have never heard of them. 

Hammu Qiyu and Sidi Hammu 

According to some reports, ‘A'isha Qandisha is married to a jinn 
called Hammu Qiyu. Like most of the male jnun, his character is 
relatively undeveloped; however, the Hamadsha insist that he must 
be distinguished from still another jinn called Sidi Hammu. Sidi 
Hammu likes red jawi and is particularly fond of blood. In fact, he 
lives around slaughterhouses and slaughter blocks, and is said to have 
a particularly intimate relationship with butchers. He has a number of 
followers among the Hamadsha, and is known to have sexual relations 
with some of the female devotees. The words of his rih are: 

Sidi Hammu, forgive us. 

O, Our Prophet, forgive us. 

Messenger of God, forgive us. 

Men of God, forgive us. 

Lalla Malika 

Lalla Malika, the daughter of the king, has the most elaborate per¬ 
sonality of the remaining she-demons popularly believed in the Meknes 
area. She does not respond to the hadra of the Hamadsha—she re- 



THE THEORY OF THERAPY 


147 


sponds instead to the dances of the Gnawa and Jilala—but she is often 
referred to by Hamadsha and has relations with many of them. Lalla 
Malika is very beautiful and dresses, as they say, with a lot of chic. 
She demands the same elegance of all her followers. She is a flirt and 
quite promiscuous, and she especially enjoys relationships with married 
men. I have been told she speaks only French and that she lives in 
clothes cabinets. 

When Lalla Malika sees someone she likes, she goes up to him and passes 
her hand in front of his eyes. The person then sees only clouds. He sees 
nothing but Lalla Malika. He cannot remember his house. Lalla Malika re¬ 
mains next to him. When she sees him standing off to the side, she calls him 
by name and tells him that she is next to him and asks him to accept her 
wishes. The man agrees. “I want you to marry me,” she says. “I want you 
to sleep with me.” The man agrees, say, on condition that he has the liberty 
to have intercourse with other women. Lalla Malika agrees. Once they have 
come to an agreement, the man asks to see her. (He has not seen her up to 
this time.) She appears in front of him with clothes embroidered in gold, 
sometimes with long hair, sometimes with short hair. Then she says all right. 
The man agrees. “I’m in your hands,” he says. He can then marry her or 
sleep with her. It is all the same. She tells him to wear perfumes, to shave 
often, and to wear new clothes. “Then you’ll have your liberty,” she tells 
him. “I’ll even help you escape prison.” 11 

Lalla Malika is always gay, and she does not attack her followers. She 
likes to laugh and tickle them, and she is responsible when a group of 
women suddenly begin giggling. She prefers the color pink, and re¬ 
quires her followers to wear eau de cologne and bum santal (/-W 
qamari). Her rih is usually played by the Gnawa or Jilala, but it may 
be played by the Hamadsha if, as happens occasionally, she should 
take possession of one of the dancers in their hadra. The words of her 
rih are as follows: 

Welcome, Lalla Malika! Welcome, O ‘Alawiyya! 

Welcome, Lalla Malika, to the hadra of Lalla Malika. 

Allah! Lalla Malika! I bcsccch Lalla Malika. 

Allah! Lalla Malika! He who beseeches can have no fear. 

Allah! Daughter of the Prophet. 

Lalla Mira 

After £ A'isha Qandisha, Lalla Mira has perhaps the greatest following 
among the women of Meknes. She makes them wear bright yellow or 

11 This example was given by the same informant who told me of ‘Aisha’s ways. 



148 


THE HAMADSHA 


orange and burn yellow jawi. I have been told that she comes from 
the Beni Mtir, a Berber tribe, and she is often called Lalla Mira al- 
Mtiriyya. She lives in houses and takes her walks after afternoon pray¬ 
ers. Her character is poorly developed. She makes people laugh, but 
also attacks and takes possession of them. She is unmarried. 

Sometimes a woman visits another and begins to gossip. Lalla Mira listens 
to them. If one of the women should criticize her, then she will attack her. 
Then the woman must wear yellow, the color of Lalla Mira, and perform 
the hadra. Lalla Mira can attack someone who is laughing or crying a lot. 
Sometimes a woman who is crying is suddenly paralyzed. She continues to 
cry as long as she is paralyzed. You must then put henna in her hands, in 
her mouth, and in her nose. She may recover or waste away. Lalla Mira also 
attacks women who do not perform the hadra correcdy. 

She responds to the hadras of the Hamadsha and the Jilala. Her rih is: 

Allah! Lalla Mira! O, Gnawa! O, my eye! 

I am your loyal servant. 

I wish God to grant me my desires. 

O, Our Prophet, forgive me. 

O virtuous saints, forgive me. 


Lalla Mimuna 

Lalla Mimuna, some say, comes from the area of the Guerouan, a 
Berber tribe to the south of Meknes. She lives in wells and old deserted 
houses. She makes her followers wear blue and bum hasalban, and 
has been likened to “a woman who does not get on with her husband.” 
She often attacks both men and women. She responds to the hadras 
of the Gnawa and the Hamadsha. I have been told that she likes to 
seduce men when they are traveling. 

Other Named-Jnim 

The female jnun play a much more active role in the lives of the 
Hamadsha than do the male jnun. The male jnun are less developed 
as characters, and their demands are not so clearly formalized. Sidi 
Mimun, Moulay Brahim (who is sometimes identified with Moulay 
Brahim, a famous saint buried near Marrakech), Sidi Musa, and Sidi 
Hammu are all important to the Meknes Hamadsha. Their rihs are 
played at nearly every hadra. 12 In other parts of the country, other 

“The words of the rih of these three jnun arc: 



THE THEORY OF THERAPY 


149 


jnun play important roles in the Hamadsha ceremonies. They are 
known to the Hamadsha of Meknes, but they have neither influenced, 
attacked, nor taken possession of them. 

The Problem of Diagnostics 

The Hamadsha—and other Moroccans of similar background—em¬ 
ploy a not entirely consistent taxonomy of diseases which appears to 
be based on three principal factors: etiology (pathogenesis), symp¬ 
tomatology, and responsiveness to cure. These factors are interrelated, 
and no attempt has been made to isolate any one of them or give it 
priority in the construction of a consistent disease taxonomy. There 
is little interest in disease classification per se. What is important is 
the efficacy of the cure. The Moroccans are pragmatists. They operate 
on the basis of hunches—hypotheses—the validity of which is con¬ 
firmed by the efficacy of the cure. A man falls ill. His general symp¬ 
toms, both physical and mental, as well as certain social considerations 
which would receive scant attention in Western medical therapy— 
Does he have an enemy? Is he better off than the other members of 
his social group? Has he more than one wife?—suggest a cause for 
the disease. The man may have been poisoned; he may be the victim 
of the evil eye; he may be suffering the effects of witchcraft. For each 
etiological hunch, a certain remedy or series of remedies is appropriate. 
If one of them proves successful, then the hunch has been validated, 
and the disease appropriately interpreted and classified. It should be 
pointed out that each of these hunches “formulates” the disease. This 
is, of course, particularly true of those diseases which the Westerner 
would consider psychogenic, but it affects at least the psychological 
dimensions of all illnesses. 

Sidi Mimun 

Allah! O Sidi Mimun the Gnawi! O my eye! 

Pardon, O our Prophet. O messenger of God, pardon. 

Sidi Mimun, pardon. The virtuous of the Sahara, pardon. 

Prophet of Allah, pardon. Messenger of God, pardon. 

Sidi Musa 

The saint! Sidi Musa! 

Pardon, Sidi Musa. O, Allah! 

O, Door of Musa in the Gnawa. 

Sidi Musa! 

Moulay Brahim 

O, Moulay Brahim, pardon. 

O, Moulay Hasan, pardon. 

As well as the virtuous men of Allah, pardon. 

I have come as a guest, pardon. 



150 


THE HAMADSHA 


Symptomatology, however, is by no means insignificant. It provides 
a limit to the range of possible etiological hypotheses. A man whose 
lips turn blue after drinking a glass of mint tea and who is then unable 
to retain any food is not to be confused with a man who suddenly 
falls into a convulsion and awakes to find the left side of his body 
paralyzed. The blue lips immediately suggest poisoning and a whole 
range of cures. The paralysis indicates that the man has been struck 
by a jinn, and suggests another series of cures. Social considerations, 
which are not necessarily articulated or even conscious, may en¬ 
courage a particular explanation or even limit it. Thus, if the hemi- 
plegiac is known to have prevented his son from attending Hamadsha 
performances, it is reasonable to assume that c Aisha Qandisha is re¬ 
sponsible for his paralysis. The range of possible cures is limited 
thereby to a pilgrimage to Beni Rachid or an invitation to the Hamad¬ 
sha. If the hemiplegiac recovers after having tried either or both of 
these cures, it is assumed that £ Aisha did in fact strike him. If the cures 
prove ineffective, a new hypothesis is formulated and new cures are 
tried. 

Symptomatology provides, then, diagnostic guidelines. Diseases, 
however, are classified less on the basis of symptoms than on etiological 
considerations, and the correctness of these considerations is validated 
by the efficacy of the appropriate cure. Since etiology provides the 
key to disease classification, its underlying structure has to be un¬ 
covered. Here I will limit my analysis to those illnesses which are said 
to be caused by the jnun. 

The jnun are held accountable for a wide variety of diseases that 
the Westerner would consider physical, and others that he would 
consider psychological. The illnesses for which the jnun are etiologic 
agents fall into two principal categories. In one, the jnun are said to 
have caused the illness, but they do not play a role in its treatment. In 
the other, the jnun are also said to have caused the illness, but here 
they are explicitly involved in the treatment of it. These categories 
are not distinguished by the Moroccan. I have chosen to term the 
particular way in which the jnun are held accountable for the first 
category of illness the explicative mode of responsibility; the second 
category, the participational mode of responsibility. 

The Explicative Mode of Responsibility 

The jnun serve simply as an explanatory device for phenomena 
which cannot otherwise be explained within the cultural system as a 
whole—or, more specifically, within the individual’s own cultural 



THE THEORY OF THERAPY 


151 


reality. The individual does not immediately associate the illness with 
the jnun. He refers to the jnun only when he is hard-pressed to find 
an explanation for the illness. For example, in discussing the reasons 
why a man goes crazy (lpamiq), one of my informants told me that 
a man becomes crazy when he has lost all of his money or all of his 
children in some disaster. I then asked him about a particular man who 
was considered crazy. This man had neither lost all of his money nor 
all of his children, nor suffered any similar disastrous experience. 14 Ah,” 
my informant said, “it is those people there.” He assured me that he 
had no idea why “those people there” chose to make the man crazy. 
It was written. Similarly, he had no idea of how the jnun made the man 
crazy. There was nothing to be done, he said. The Hamadsha, the 
Gnawa, or even the Koranic teachers who specialize in exorcism could 
do nothing. True, the man’s family could take him to a saint’s tomb; 
there was always the chance that a saint could help him, but this was 
doubtful. The man was sure to end up in Berechid, the government’s 
mental hospital. 13 

The jnun function in the explicative mode of responsibility as a 
full stop in the speculations of an individual. They may serve to ex¬ 
plain why or even how that individual fell ill. They set a limit to the 
causal chain; there is nothing more to add but to say that it is written, 
that it is the will of God. In this mode of responsibility, the jnun do 
not represent, reflect, or symbolize a dynamic process within the 
disease. They are, rather, a logical construct whose symbolic function 
is much more diffuse. 

The Participational Mode of Responsibility 

Here, too, the jnun function as an explanatory device for phe¬ 
nomena which cannot otherwise be explained either culturally or per¬ 
sonally. No other explanations will be offered first, however; the 
jnun do not serve in this case as a full stop to speculation about the 
illness. The illness is related immediately and dynamically to the 
jnun. Not only have the jnun—or, more often, a single jinn—caused 
the illness, but they “participate” somehow in the illness itself and must 
be appeased to bring about a cure. In this mode of responsibility the 
jnun are more than a logical construct. They represent, image, or sym¬ 
bolize an etiological factor or a dynamic process within the illness it- 

“When a man goes crazy because he has lost all his money or all his children, 
it is usually not the jnun who are responsible for these disasters. They are at¬ 
tributed to the evil eye—to the jealousy of someone within the man’s society. 



152 


THE HAMADSHA 


self (see below). It is precisely with this category of diseases that the 
Hamadsha are concerned. 

A Taxonomy of Jinn-Produced Illnesses 

The Hamadsha—I am restricting myself here to the Hamadsha in 
and around Aleknes, since there are considerable dialectical differences 
in Morocco—recognize several different ways in which the jnun can 
cause an illness in their victims. 14 The mode of attack is usually sug¬ 
gested by the symptomatology. The person who exhibits the symp¬ 
toms of a possession state is said to be “inhabited” (inaskuri) by a jinn, 
whereas the person who suddenly is paralyzed is said to have been 
struck ( madrub) by a jinn. These distinctions in attack, however, only 
partially correspond to symptomatic differences. Thus, while it is 
possible to differentiate linguistically between a blind man and a par¬ 
alyzed man, it is also possible to refer to both of them as madrub , 
as “having received a severe blow by a jinn,” if both the blindness and 
paralysis occurred under certain circumstances—suddenness of onset, 
for example—and if the men exhibit no counterindicative symptoms— 
such as, in the case of the blind man, a cataract. In only one instance 
is a category of the jnun-produced illnesses symptom-specific: matrush. 
The word matrush refers to a sudden, unilateral paralysis of the face, 
usually on the left side. It is believed to be the result of a slap in 
the face by a jinn. 15 

The jnun attack in two principal ways. They may strike a person, 
or they may take possession of him. If the victim is struck, he usually 
suffers from some detrimental bodily change: blindness, deafness, 
mutism, paralysis of the limbs or face. If the victim is possessed, he 
exhibits all of the classical symptoms of “possession states”: falling un¬ 
conscious; syncope; convulsions; tremors; speaking in tongues; sud¬ 
den and abrupt, often meaningless, changes in conversation or activity; 
flights of thought; and so forth. The jnun are believed to enter the 
body of both the possessed and the struck. They remain with the pos¬ 
sessed; with the struck they enter and leave almost immediately. 16 

While the terms describing individuals who have been attacked by 

14 Although there is a difference in terminology, I do not believe that the 
basic distinction between being struck and being possessed is ignored in any 
dialects. 

“It should be noted that the Moroccan Arab is particularly sensitive to slaps, 
which he finds far more insulting and humiliating than blows of the fist. 

18 Just as possession states must be distinguished from states of being struck, so 
must they be distinguished from trance states (Bourgignon 1968; Wallace 1959). 
Possession is an interpretation of a syndrome which can occur in everyday life 
or during the trance dance. All trances are not interpreted as possession states, as 



THE THEORY OF THERAPY 


153 


the jnun are often used loosely to refer to either or both modes of 
attack, they do suggest the mode of attack and qualify its intensity. 
They may be arranged as shown in Figure 8. 


MAJNUN 



STRUCK 


MUSHAR 

(MAWALI) 

(MATRUSH) 

MADRUB 


Majnuji appears to be used most often as a generic term for anyone 
who has been attacked by a jinn. It may refer either to someone who 
has been struck or to someone who is possessed. When used more 
specifically, it refers to someone who is possessed, and not struck. 

Masift and imklif are descriptive of a person who is temporarily 
possessed by a jnun but not in need of any treatment. MasuH refers 
to someone obsessed with an idea; it may be translated as monomania- 
cal. So long as the jinn is in the person, that person is obsessed with 
an idea of activity. Makhltf refers to someone who is afraid, transfixed 
with fear, or even paralyzed with fear. It is the jinn who, entering 
him, has caused this state of fear ( khal'a ) (Westermarck 1926(1): 
276). These terms are used to describe sudden and unexpected moods 
which come over a person and for which he does not feel himself 
fully responsible. References to the jnun in explaining these and other 
similar states are frequent. 17 Although the person who is either masu't 

we shall see. Possession in everyday life, which may involve trance-like (disso¬ 
ciative) states, is considered an illness. During the trance dance, it is considered 
part of the course and of therapeutic import. As a given in the Hamadsha’s world, 
possession serves not only as an interpretation of a particular syndrome but also 
as a model for behavior. 

"Compare this function of the jnun with “the intervention of nameless and 
indeterminate daemon or ‘god’ or 'gods’” in the Iliad (Dodds 1966:11). 



154 


THE HAMADSHA 


or makhlu' does not usually require the hadra of the Hamadsha, both 
terms are often used to describe the prodromal condition of a man 
who later becomes possessed or is struck and requires the hadra or 
some other treatment. The Hamadsha, particularly those who trance 
frequently, will usually feel the need to perform the hadra after ex¬ 
periencing such states of fear or obsession. 

Mary alp is used to describe a person who is possessed. 

Maskun is more specific than 7naryah> since it refers to an individual 
who is actually inhabited by a jinn. It is the strongest word that can 
be used to describe someone possessed. Here the jinn is well lodged in 
the person, who exhibits the classical symptoms of possession. 

The maskunin [pi. of maskwi\ talk with you normally and then suddenly 
they appear to be in flight. Sometimes they talk and agree with what you 
are saying, then suddenly they will not agree with anything. Sometimes 
their brain changes. Sometimes there is something [the jnun] that takes 
hold of their brain and their head turns. Then they are all right again. They 
only felt something pinching their brain. 18 

The Hamadsha consider sudden changes in conversation or activity, as 
well as convulsions, to be the significant symptoms of a person called 
maskun. 

Maqius is the term most frequently used to describe a person who 
is attacked by the jnun. It may refer both to someone who is possessed 
by a jinn and to someone struck by a jinn. The jinn is usually con¬ 
sidered to be in the person. The person may also suffer some bodily 
change. One of my informants explained the term to me as follows. 

A person is maqius by the jnun. He feels absent [ghaib']. It is as though 
someone is in front of the person, and the person is outside his body. Some¬ 
times he falls and trembles. [I asked him where the jinn is.] God only knows 
where the jinn is. It is possible that the jinn is in the body. But this is not 
sure. There are some who are ghaib and tremble and fall, and others who 
are sick and remain seated and immobile in the comer [mqallesh], 

Ghaib is often used to describe the condition of a person in trance. 
It is usually translated as “absent” but, as my informant seems to sug¬ 
gest, it refers to a state of dissociation. 

Another of my informants explained maqius this way. 

“This description and the one that follows were given to me by an older 
Hamdushi who has been involved in the treatment of many of the devil-struck 
and the devil-possessed. He himself has been maqius on more than one occasion. 
His descriptions were confirmed by many of my other informants. 



THE THEORY OF THERAPY 


155 


There are people who are walking normally and suddenly their foot stops. 
They cannot move it. It always happens to people who are crossing a 
stream and to not say bismillah r-rahmcm r-rahim [in the name of Allah 
the merciful and the compassionate]. When you say this the jnun will not 
approach you. Only one of the man’s legs is paralyzed—the one that steps 
on the other side of the stream first. A hand can be maqius if you throw a 
rock and hit a jinn by accident. The jinn is only in the part of the body 
where the person is sick. He can talk normally. The jinn is wounded in the 
same place. The person is not maskun. 

A person who is maqius may perhaps best be described as in a state 
of partial possession. 

Mushar is derived from the verb shir, to point, to point a finger at, 
to signal. It is the weakest of all the expressions used to qualify a 
person struck by a jinn. I have been told that the jinn does not even 
graze his victim; he simply points at him. The victim seldom bears a 
mark of the jinn—manifest bodily harm or malfunctioning. At most, 
he undergoes a slight, often temporary paralysis. The more usual 
symptom is general malaise. The term often refers to the very first 
contact a person ever has with a jinn. 

Matnish has already been considered. 

Madmb , derived from the verb to beat ( darb ), qualifies a person 
who has been severely struck by a jinn. This usually involves paralysis 
of the legs, feet, arms, or hands—most frequently on the left side—tut 
it may also refer to sudden blindness, deafness, or mutism. “The person 
has an arm or leg that does not work. He cannot clench his fist quickly. 
Sometimes you see someone who has a string around his slippers to hold 
them on. Such a person is madrub.” “To be madrub is more dangerous 
than to be maskun. It is stronger. The moment a person who is madrub 
falls, you think he will die. He can fall off a tree or a mountain.” 

All of these terms, in one grammatical form or another, are used 
frequently by Meknes Hamadsha to describe victims of the jnun. Oc¬ 
casionally, they refer to these victims as slaves ( mamlukin) of the jnun. 
There are, however, many other ways of referring to the victims, to 
their condition, and to their relationship with the jnun. Often these 
involve complex paraphrases. 19 

"The most common expressions—many of which do not indicate so neady 
whether die victim is struck or possessed—are as follows: fih l-aryah (1-ary ah are 
in him); fib ej-jnun (the jnun are in him); bib ej-jnun (the jnun are with him); 
shadduh l-aryah (l-aryah have seized him); and tiyuh l-a'dam or tab bi- 
l-a c dam (fall into epileptoid convulsions). 

It should be noted that a c dam may mean bones. The relationship between bones 
and attacks by the jnun and the need to trance is discussed later. 



156 


THE HAMADSHA 


There is considerably more fluidity in the use of the terms to describe 
the jinn-struck and the jinn-possessed than my paradigm suggests. The 
fluidity stems not only from the normal looseness of everyday speech 
but also from the type of relationship the jnun maintain with human 
beings. This is subject to considerable variation. Often a person who 
first is struck ( madrub) will later become possessed ( maskun ). More¬ 
over, these terms are used to describe various conditions of an individ¬ 
ual in trance who may be possessed, paralyzed, or rendered blind, deaf, 
or mute by a jinn. One Hamdushi put it this way, after we had dis¬ 
cussed the meaning of the words in my paradigm: 

All of these words can be used in the same way. To be madrub is to have 
been given a good blow. Maqius, a much lighter one. Matrush is from a 
slap. Maskun occurs when the jnun are in the body. Sometimes when the 
maskun talks he is all right, and at other times he confuses everything. A 
person is mushar when the jnun have reached out but have not touched the 
person they want to slap. It is the least strong. Muwali is a light touch. 
[I have not discussed muwali in my paradigm, because it is more often used 
to describe a relationship with a jinn or jiniyya.] A person who is madrub 
may be distinguished from a person who is maqius by the fact that he 
(the former) always has a mark—a black-and-blue mark, a scratch, or a 
swelling, paralysis, drooling, a tight mouth. The person who is ?naqius is 
always sick, but he does not know where. Matrush is always the face. 
Madrub refers to other parts of the body. Maskun refers to a person who 
is inhabited for a long time. Muwali, just struck. Mushar, never before 
struck, just touched. 

Diseases of the Named and Unnamed Jnun 

The Hamadsha and other Moroccans of their background do not 
distinguish between illnesses caused by the named-jnun and those 
caused by the unnamed ones, just as they do not distinguish between 
named and unnamed jnun. However, in their etiological theories there 
seems to be a difference between the two types of illness, and these 
differences are reflected in their choice of treatment. (There is not, to 
my knowledge, any difference in manifest symptoms.) It has to be re¬ 
membered, nevertheless, that the distinction between diseases caused by 
named-jnun and those caused by unnamed ones is analytical and as such 
implies a certain consistency and rigidity which is not found in as prag¬ 
matic a therapy as the Hamadsha’s. 

While it is possible for the jnun, both named and unnamed, to attack 
a human being for no apparent reason, it is rare. Most illnesses for 
which the jnun are held accountable arc the results of transgressions 



THE THEORY OF THERAPY 


157 


committed by the sufferer against the jnun. Both the named and the 
unnamed jnun, being sensitive to insult and injury, are quick to re¬ 
taliate for wrongs committed against them. In the case of most illnesses 
for which the unnamed jnun are held accountable, the victim is un¬ 
aware of his transgression. Often it is suggested that the victim inad¬ 
vertently trampled on a jinn while walking at night or that he threw 
boiling water or fire on it, particularly if he threw the water or fire 
into a drain, a spot popular with the jnun. 

The initial attack by a named-jinn—attack is one of the most com¬ 
mon ways a named-jinn establishes a relationship with a human being 
—resembles an attack by an unnamed jinn. Often, the victim does not 
know how he has offended the jinn. Speculations about the cause of the 
illness are similar to those for illnesses caused by the unnamed jinn. 
Once the name of the jinn is uncovered (see below), and the victim 
has been appropriately treated, most, if not all, subsequent “attacks” 
are interpreted as an attack by that particular jinn. The victim is con¬ 
sidered to be dependent (: rnuwali, muttakil) on the jinn from that time 
on. If the victim, without ever having been previously attacked, is 
already dependent on a named-jinn, symptoms of an attack by the jnun 
are also interpreted in terms of the jinn he follows. This son of de¬ 
pendency relationship is formed most often during the hadra, when a 
person who is not ill is thrown into trance by a rih known to be the 
favorite of a particular jinn. 

Attacks by a named-jinn on someone already dependent are inter¬ 
preted as the result of some act that incurred the rage of the jinn. 
Most often the jinn’s rage is said to be brought on by his follower’s 
failure to obey one of his commands. The follower may have neglected 
to wear the appropriate color or to have burned the appropriate incense 
on schedule. He may have failed to sponsor a promised ceremony, or 
he may have tried to postpone a sacrifice. More generally, he may not 
have lived up to the standards of propriety expected of the followers 
of a jinn. The point to be emphasized is that the follower had dis¬ 
pleased his jinn and “the jinn—they are not like human beings—do not 
go in for any nonsense.” 

Therapeutic Theory 

There are a number of possible remedies for a person who is struck 
by a jinn. The purpose of these remedies, according to Westermarck 
(1926(I): 324), is to drive the enemy out. In fact, this is only partially 
true. There seem to be two basic approaches to the cure of victims of 



158 


THE HAMADSHA 


the jnun. One of these is to drive the jinn out permanently, but the 
other is to establish some sort of a “working relationship” with him. 
The first approach—exorcism, properly speaking—is considered by the 
French classicist Jeanmaire as characteristic of those countries in which 
the possessing spirit is considered evil; the second is to be found in 
North Africa, the Sudan, and among the ancient Greeks. 

II semble qu’on puisse valablement caracteriser le processus en observant 
qu’a la difference de l’exorcisme proprement dit, tel qu’il se pratique dans 
un milieu ofi la possession est comprise comme l’effet de l’intrusion d’un 
esprit maleficient par nature et oil l’exorcisme tend, par suite, a l’expulsion 
de ce principe mauvais pour produire la delivrance du sujet, la methode de 
traitement appliquee ici [in North Africa and the Sudan] vise moins a la 
suppression des etats affecrifs et delirants qui resultent de l’etat de posses¬ 
sion (suppose ou suggere) qu’a leur transformation par elimination du fac- 
teur depressif et a leur utilisation en vue de realiser un equilibre nouveau 
par une sorte de symbiose avec l’esprit possesseur devenu esprit protecteur 
et par une normalisation, sous forme de transe provoquee, de 1’etat de crise. 
(1951:121; quoted from Jeanmaire 1949) 

In Morocco, both exorcism and the establishment of a symbiotic re¬ 
lationship with a jinn are popular. Although no hard and fast rule can 
be established, it appears that most often exorcism ( c azima , lit. incan¬ 
tation) is employed when a person is attacked by an unnamed jinn, 20 
and the establishment of a symbiotic relationship is effected when one 
is attacked by a named-jinn (cf. Westermarck 1926(1): 155-156). I 
would suggest, further, that the two modes of possession and their treat¬ 
ment correspond to different psychic states. Possession by an unnamed 
jinn is probably representative of one psychic condition or dynamic 
process, and possession by a named-jinn is representative of another. 

While the Hamadsha, as well as the Gnawa, Jilala, and adepts of 
similar confraternities are sometimes called to treat the victims of un¬ 
named jnun, and while they are often successful in such cases, they are 
primarily concerned with the treatment of victims of named-jnun. 
They establish or reestablish a working relationship between the patient 
and his jinn. I do not mean to imply that the Hamadsha consciously 
decide to treat one man because he is known to be the victim of a 
named-jinn and not to treat another because he was attacked by an 
unnamed jinn. Rather, the Hamadsha are called on to treat a patient 
because it is believed that that patient was attacked by ‘A'isha Qandisha, 

"The verb sara c , to overcome, to throw to the ground, means also to exorcise. 
Its basic meaning is indicative of the permanent expulsion of the jinn in exorcism. 




THE THEORY OF THERAPY 


159 


Sidi Hammu, or some other jinn over whom the Hamadsha exercise 
control. Sometimes the Hamadsha are invited to perform when it is 
not certain who attacked the patient. Since a hadra is expensive, these 
trial and error cures are quite infrequent. When they do occur, there 
are invariably other causes for the invitation. Even in cases where the 
jinn is known to be ‘A'isha Qandisha or Sidi Hammu, other, cheaper 
cures such as a pilgrimage or attendance at a public ceremony are 
sometimes tried first and found to be successful. It is my impression 
that often such cures are not of long duration. The patient will have a 
relapse, and this ultimately necessitates an invitation to one of the 
brotherhoods. 

The major difference between exorcistic and symbiotic cures is that 
exorcistic cures are one-shot affairs and symbiotic cures are continuous. 
The patient, in the case of a symbiotic cure, is incorporated into a cult, 
and, as a member of that cult he must go through “curing” periodically. 
The exorcistic cures are much simpler. They may involve nothing 
more than placing a little tar in the victim’s nostrils and other body 
orifices, bathing him in water endowed with baraka, or burning incense 
over him. Often, a Koranic teacher or an exorcist will incant verses 
from the Koran over the victim until the jinn leaves him. Sometimes 
sacrifices are required. These sacrifices may be conceived either as a 
gift ( hadiyya ) to the jinn or as an c ar —literally, shame; “an act which 
intrinsically implies the transference of a conditional curse for the pur¬ 
pose of compelling somebody to grant a request” (Westermarck 
1926(1):518). Or they may involve the ritualized preparation of special 
food such as a chicken cooked without salt ( viassiis ), which is fed to 
the patient (that is, to the jinn within him). Both sacrifices and the eat¬ 
ing of special foods ( dyafa) are not restricted to exorcistic cures; they 
are often part of the symbiotic cures as well. 


Symbiotic Cures 

The first step in the symbiotic cure is to establish the identity of the 
jinn who has struck or possessed the patient. This may be accomplished 
informally, if there is some evidence that a particular jinn has struck 
the patient. The evidence is often circumstantial. 

Q. was about 14 years old when he was first struck by a jinn [ madrub ]. 
His family had recently moved to Aloulay Idriss to find work and were 
staying with his sister. Q. laughed at a group of Sidi 'Ali’s followers who 
were performing in the street. He claims that he had never heard the 



160 


THE HAMADSHA 


Hamadsha before. The moment he laughed, he was stricken with a paralysis 
of the lower limbs. His sister immediately understood that c AIsha Qandisha 
had struck him, and invited in the Hamadsha to perform the hadra for her 
brother. First a sacrifice—probably a goat—was made, and then the Hamad¬ 
sha danced for Q. Q. too began to dance and to slash his head. He has 
been a follower of Sidi c Ali ever since, and considers himself to be de¬ 
pendent [mu'wali'] on ‘Aisha Qandisha. 

Identification may also result from a dream at the time of the jinn’s 
attack or shortly afterward. 

ill., who had been a member of the Miliana, a brotherhood whose members 
play with and eat fire, reported the following dream which converted him 
to the Hamadsha and was responsible for his becoming a follower of c AIsha 
Qandisha. “I was sleeping. Lalla c Aisha grabbed me by the neck and threw 
me to the ground. This happened three rimes. [M. had the dream on three 
consecutive nights.] When I changed over to the Hamadsha, I got better. 
‘Aisha told me, ‘Either you work for me, or I’ll have your neck.’ ‘God is 
my witness,’ I cried out with all my might. But there was nothing to be 
done. I was sick the next day. My whole body was sick. I could not walk. 
My wife brought me this black chemise and this red turban. [These are the 
colors of c Aisha Qandisha.] Then I felt better.” M. danced the hadra on the 
third day and slashed his head. “ c AJsha asked me to work for her [per¬ 
form the hadra] for the rest of my life. I will always work for Lalla 
c Aisha.” 

Often the victim of a jinn travels to a saint’s tomb and waits there for 
a dream which will reveal the cause of his illness. 

If there is neither circumstantial evidence nor oneiric evidence for 
the identity of the jinn, the patient may be taken to the ceremonies of 
the Gnawa, the Hamadsha, the Jilala, or other popular brotherhoods— 
ceremonies held for other people—to see whether or not he responds to 
their music. This is one of the most common diagnostic methods in the 
shantytowns of Meknes, where there are ceremonies almost nightly. 
Attempts to identify the patient’s jinn are often made during these 
ceremonies; his ear is pulled until the jinn speaks. Here, suggestions 
play an important role. 

F., a girl of 7 or 8, was struck in the tongue by a jinn, so that every time 
she wanted to talk her tongue would roll back and she would remain speech¬ 
less. It was thought that ‘Aisha Qandisha was perhaps responsible for this, 
and F. was taken to a Hamadsha ceremony given by a neighbor who had 
been blinded by c Aisha. F. was placed in a small room opening out onto the 



THE THEORY OF THERAPY 


161 


court where the Hamadsha were performing and was generally ignored. 
After the Hamadsha had performed for almost an hour, the muqaddim came 
in to see how she was doing. (He was excited but not in trance.) F. had 
not responded to any of the rihs that had been played. 

Following is an excerpt from my notes, taken at the time: F. looked at 
me curiously. Her eyes were wide-open, popping. Her focus seemed out of 
control; her pupils rolled from side to side. Her gestures, especially with 
her hands, were clumsy: she would reach out too far in front of her for an 
object. She seemed disoriented. Her body and eye movements were “sexy,” 
in the way hysterics tend to be sexy. (I found this surprising in such a 
young girl.) F. would occasionally lean toward one of several other girls 
of about her own age who were seated near her, and rub cheeks with her; 
she would then kiss the girl, or slap her, or both—there was no way of 
predicting which. The little girl’s mother, who was watching, would push 
F. away when she slapped her daughter, but would ignore F. otherwise. 

The muqaddim came in unexpectedly. Suddenly he pulled F.’s ears very 
hard. She was pained and resisted a little. He cried out: “Leave her! Leave 
her! Leave her alone! Let her be free!” The muqaddim then banged his head 
against the terrified child, whose eyes were popping out. He had not 
banged her head very hard; it was more of a rubbing movement. He then 
spit into her mouth several times. He crossed her arms and squeezed her 
fingers tightly until it hurt her. He cried out again: “Leave! Leave, leave, 
leave her! Let her be free!” F. sat back terrified and wide-eyed. She looked 
at me and the others uncomprehendingly, her pupils rolling around. Oc¬ 
casionally her hands tightened spasmodically; I was told that from time to 
time her hands had become paralyzed before she was brought to the cere¬ 
mony. The muqaddim returned to the dance floor as abruptly as he had 
appeared, and F. sat back, ignored, for a few minutes. Ten minutes after 
the muqaddim’s two-minute treatment she got up and walked out into the 
court to watch the dancing. Then she disappeared and was not seen again. 
I imagine she had gone to sleep. 

F. did not dance and had not shown any particular response to the music 
of the Hamadsha. A few days later I met her mother and was told that 
she was in much the same condition. Two months later I saw the mother 
again and was told that F. was better. She had received no further treat¬ 
ment in the interim. Even though she had not responded to any of the 
rihs, e Aisha Qandisha was considered responsible for her illness, and it was 
not unreasonable to expect that F. would become a devotee of the Ha¬ 
madsha and a follower of the camel-footed she-demon. 

Sometimes the victim of a jinn is taken to the muqaddim of one of 
the local brotherhoods, who will try to discover the jinn’s identity by 
reading or chanting words from the Koran over him. Incense may be 
burned. Often the muqaddim will yank at the patient’s ears or, if he 



162 


THE HA MADS HA 


is a Hamdushi, will knock his head against the patient’s head. Once 
slashed, the head of a Hamdushi is considered to be greatly endowed 
with baraka and to have powerful influence over the jnun. The muqad- 
dim also may take the patient to the place where he was first attacked, 
and he may even sacrifice a chicken or some other sacrificial victim 
there. When the patient sees the blood of the sacrifice, he—or his jinn 
—becomes excited and begins to speak. 

Since the ceremonies are expensive, shorter trial performances are 
sometimes given, especially if no regular ceremonies are scheduled in 
the neighborhood. In the countryside, where the ceremonies are less 
frequent, one of the musicians may visit the patient and play various 
rihs on the ganbri to find the rih to which the patient responds. 

H. was struck by a jinn. A musician who played the ganbri was invited to 
her house and played all of the rihs he knew over her until he learned 
which one was to the liking of her jinn. This took nearly three days. When 
the appropriate rih was found, the Hamadsha were called in to perform the 
hadra. 

One of the commoner ways of identifying an attacking jinn is to call 
in a talla\ or exorcist-seer, many of whom are women. 21 Although 
these seers appear to be most closely associated with the Gnawa and 
the Jilala—for example, a talk' must sponsor a ceremony for the Gnawa 
and another for the Jilala every year—the seers in Meknes are by now 
also associated with the Hamadsha, the most active of the brotherhoods. 
The talla' attempts to identify the jinn by either mechanical or oracular 
divinatory practices. The most popular of the mechanical methods em¬ 
ployed in Meknes involves casting cowrie shells and “reading” from 
the pattern in which they fall. I do not know to what extent the rules 
for reading the shell-patterns are formulated and to what extent the 
reading is simply a theatrical device to add authority to the seer’s words. 

In the oracular divinations, the talk' burns incense and inhales the 
fumes. Then, in what would seem to be a state of partial dissociation, 
he begins to mumble often incomprehensible words and phrases. These 
utterances are usually highly interspersed with the names of jnun and 
saints. Finally the seer stops and informs the patient that he is possessed 
of such-and-such a jinn and must follow the jinn’s commands. These may 
involve visiting a saint, burning special incense, or inviting the Gnawa, 

“Although the talla' is sometimes referred to as a shuvrwaf, he is usually dis¬ 
tinguished from the run-of-the-mill seers who are not necessarily associated with 
one or more brotherhoods. 




THE THEORY OF THERAPY 


163 


the Jilala—or, nowadays, the Hamadsha—to perform the hadra. Often 
the talla' will instruct the patient to wear certain colors. The remedies 
suggested by the talla c are stereotyped and not unfamiliar to the patient. 
It is significant that the seers usually ask no questions. They simply tell 
the patient what is wrong. Sometimes the patient will ask the talla' a 
few questions, in order to test him, before bringing up the matter at 
hand. What the patient seems to desire is an authoritative diagnosis and 
suggestion for treatment. Once again the element of suggestion is 
strong. 

It is difficult to determine whether or not these seers believe com¬ 
pletely in what they are doing. I personally had little contact with 
them and am only able to suggest that even the least sincere are, while 
performing a divination, convinced of its efficacy and its truth. They 
appear to put themselves into an extraordinary psychic state, perhaps 
one of partial dissociation, in which they give, not totally free reign 
to their associations, but limited reign. This procedure may be called 
directed free association. The extent to which a talla c in this partially 
dissociated state is more sensitive to the needs of his client remains an 
open question. 

Perhaps, phenomenologically speaking, it is irrelevant to refer to an 
individual in this state in terms of bad faith. In fact, “suspended dis¬ 
belief’ may well be one of its essential characteristics. To what extent 
the seers are sincere and have faith in their powers when they are not 
dissociated is another question, and must be answered on individual 
bases. Certainly their close ties with a team of performers, their well- 
recognized greed, and their highly streotyped responses suggest an 
element of self-interest in their behavior. Most Moroccans consider 
them charlatans but will not hesitate to use them in a time of crisis. As 
constituent elements in the total therapeutic process—a process which 
is often highly successful—the seers cannot be dismissed as entirely 
ineffective. 

Having established the identity of the attacking jinn, it is then neces¬ 
sary to discover what, in fact, the jinn wants from his victim. Often the 
jinn’s demands follow directly from its identity; the jinn is asking the 
usual, well-known things it asks of all its followers. The victim must 
wear special colors, burn special incense, and dance to particular rihs. 
Often he must make a sacrifice to the jinn and sponsor a ceremony of 
one of the popular brotherhoods. If the person is struck by 'Ai'sha 
Qandisha, a pilgrimage to Beni Rachid and Beni Ouarad, with a stop-off 
at 'Alsha’s grotto with a red or black sacrificial chicken, may be neces¬ 
sary. Frequently the pilgrimage may be followed by a sponsored hadra, 



164 


THE HAMADSHA 


either in front of Sidi c Ali’s or Sidi Ahmed’s mausoleum or, more com¬ 
monly, at home. 

In some instances, the demands of the jinn are not known. A seer or 
the muqaddim of a team who knows the jinn’s rih will be consulted. 
The talla c , who receives a kickback from the team he recommends, or 
the muqaddim who is anxious to have his team perform, will usually 
suggest that the jinn requires the hadra and a sacrifice. 

A., a boy of 8 or 9 who lives with his mother and brother in Sidi Baba, was 
struck by a jinn. He fell into a mudhole in convulsions, and when he re¬ 
covered, his right hand was partially paralyzed. The following day, while 
he was at work in the public ovens where he is an apprentice, he suddenly 
picked up some red-hot coals and, squeezing them in his hand, announced 
that it was necessary to call in the Hamadsha and that he should be taken 
to a nearby spring. He then released the coals, which were too hot for 
anyone else to touch; he had shown no signs of pain and suffered no burns. 
In the afternoon he was taken to the spring by the muqaddim of a local 
Hamadsha team. It was assumed that ‘ATsha Qandisha had struck him there. 
Although I was not present at the spring, I was told that the muqaddim 
was unsuccessful in learning the demands of the jinniyya. He evidently 
pulled the boy’s ears, rubbed mud from the spring on him and his mother, 
and knocked his head against A. It was finally decided to give a trial per¬ 
formance of the hadra. Since A.’s mother was very poor and depended on 
her son, a short hadra was performed in one of Sidi Baba’s squares. Dona¬ 
tions from all the neighbors of the child’s mother were made. These paid 
for the team, and some money was put away for the sacrifice, which was 
considered necessary. The hadra was performed, and the little boy entered 
a light trance and attempted to hit his head first with a knife which was 
taken away from him and then with his shoe. He danced off to the side and 
was generally ignored. After about five minutes the music stopped, and the 
boy fell to the ground. 

Following is a description by my field assistant of what I was unable to 
see further back in the crowd (note the way he refers to the possessed boy 
by the name of the possessing jinniyya, c AIsha): “. . . The boy’s mother 
watched the dancers. She herself did not dance. She was dressed in black. 
The sick boy danced hitting his head, I found this curious. Yesterday I saw 
his hand paralyzed, and today he is hitting his head with the very same hand. 
The hand is perfectly normal. . . . When the music stopped, the boy fell 
down. An old man took the boy by the hand and made an ahd [squeezed 
the right hand of the boy with his own in a solemn oath]. In making the 
ahd, he tried to talk to c Aisha. ‘Aisha did not want to speak. c Aisha made 
only gestures. It is the boy who makes the gestures, but it is in fact c ATsha 
who viakes them. The old man asked c ATsha what she wanted. ‘If you want 
a sacrifice,’ he commanded, ‘tell me what kind you want.’ ‘Alsha beat her 



THE THEORY OF THERAPY 


165 


chest. ‘Leave the little boy!’ the old man commanded. ‘Why do you attack 
him? Why don’t you attack someone else?’ c AIsha beat her chest again. The 
old man said, ‘I ask you in the name of God and Sidi c Ali to tell me what 
you want. Or leave forever.’ c Aisha said no. She moved her hand across her 
chest. ‘If you want a sacrifice,’ the man continued, ‘do not hesitate. We are 
ready to make it.’ The boy said no. The Hamadsha burned more incense. 
They moved the [right] hand of the sick boy over the brasier. Then the 
man ordered the boy to be taken to the public oven. I do not think the 
jinniyya accepted the sacrifice.” 

Before the boy was taken to the oven, one of the team members, who 
claims to be a sharif, a descendant of the Prophet, spit into the boy’s mouth 
several times, massaged his hands, and said a prayer ( fatha) over him. 
I believe that there was more mention of a sacrifice than my field assistant 
reported. One of the men distinctly suggested a he-goat. The boy himself 
was in a dreamy, semi-conscious state. Occasionally he would beat his hand 
against his chest. Everyone seemed eager to accept this as an indication 
of ‘Alsha’s will. (It should be noted that a few months earlier, A.’s brother 
was attacked by c ATsha and had a performance of the Hamadsha sponsored 
for him. A. is said to have been jealous of his brother.) Several days later, 
A.’s hand was still weak, although now he could move it. It was assumed 
that he would remain in this condition until a sacrifice had been made. 
No sacrifice was made, to my knowledge, because A.’s mother was too poor. 
A.’s condition improved slightly in the following months. He appeared at 
other Hamadsha performances and danced the hadra. 22 

It is significant here that everyone concerned with the patient was sure 
of c A'fsha’s demands. The problem was simply that of ‘Aisha’s accepting 
the offering. And the question that remains is this: To what extent was 
‘Ai'sha’s “refusal” indicative of the boy’s desire for prolonged attention 
and to what extent was it indicative of the boy’s knowledge of his 
mother’s inability to make the sacrifice and of his own inability to take 
no for an answer? I was never able to establish sufficient contact with 
A. or his mother to explore and clarify this. Nevertheless, his case 
again calls attention to the role of suggestion in the diagnostic process. 

Once the identity of the attacking jinn and its demands are estab¬ 
lished, it is then necessary to satisfy these demands if the patient is to 
be cured of his symptoms. The manifest aim of the cure is the elim¬ 
ination of symptoms, and not the permanent expulsion of the jinn. The 
therapeutic process, however, is conceptualized, often confusedly, in 
terms of the jinn. The jinn must be placated, since the patient, wit¬ 
tingly or unwittingly, has incurred its wrath. This placation, then, is 

* The question of sharing expenses for a hadra is discussed in Chapter Ten. 



166 


THE HAMADSHA 


accomplished by complying with the jinn’s commands. If the patient 
is possessed or partially possessed, the elimination of symptoms neces¬ 
sarily involves the expulsion of the jinn. If the patient has been struck, 
the cure usually involves the jinn’s taking possession of the patient and 
then leaving him. In neither case is the break complete. Although the 
jinn leaves the patient, it does not abandon him. It remains with him, 
and so long as the patient obeys it, the jinn will do him no harm. If, 
however, the patient breaks any of the jinn’s commands, then the jinn 
attacks him. It appears that cure is effected by the establishment of a 
relationship between the patient and his named-jinn. The patient be¬ 
comes a follower of the jinn and remains always dependent on it. This 
relationship, as we have seen, is by no means a negative one. The jinn is 
a support for the patient. 

The Explanation of the Cure 

There are several divergent and often confusing theories as to how 
exactly the cures are brought about. Usually it is said that either Sidi 
c Ali or Sidi Ahmed or both, as intermediaries to God, are ultimately 
responsible for the cures. If the patient was struck by c A!sha Qandisha, 
it is sometimes added that Sidi c Ali commanded c Aisha to leave her 
victim alone. I have been told by the followers of Sidi c Ali that their 
saint not only commands c A'isha Qandisha but also Sidi Ahmed, and 
that sometimes he will instruct Sidi Ahmed to order ‘Ai’sha to leave the 
patient. Sidi Ahmed’s followers are not so anxious to relinquish ultimate 
control to Sidi c Ali, and they talk of an agreement ( c ar) that Sidi Ah¬ 
med made with c Aisha Qandisha, wherein c Ai‘sha agreed to keep out 
of Sidi Ahmed’s way. Sidi Ahmed, thereby, is not without his own 
effect on the camel-footed she-demon. 23 

The most frequent explanation of the cures involves the notion of 
baraka. The baraka of Sidi c Ali and Sidi Ahmed, which is lodged in the 
saints’ paraphernalia, in their descendants, and in some of their follow¬ 
ers at least some of the time, is considered to be the cause of the cures. 
In the case of the saintly paraphernalia—the tomb, the water of the 
c Ayn Kabir, and objects sold by the saints’ families—and the saints’ 
descendants, the baraka is passed directly to the patient, and in the proc¬ 
ess of transference is transformed, as we have seen, from a highly con- 

® On several occasions I have been told that the Hamadsha cure the possessed 
by allowing the possessing jinn to enter themselves. The carers, who are in a 
special relationship to the possessing jinn, are not bothered by it. This explana¬ 
tion, which resembles explanations of the Zar cures, is unusual among the Hamad¬ 
sha (cf. Leiris 1958). 



THE THEORY OF THERAPY 


167 


tagious potentializing force to a noncontagious potential or actual state 
of being. The baraka may be sufficient to eliminate the patient’s symp¬ 
toms, but it is rare in those cases where the patient has been attacked 
by a named-jinn. Usually, in such cases, the baraka seems to be trans¬ 
formed into a potential state of health which will be actualized only if 
the patient follows a regime. The regime is the command of the jinn 
or jinniyya. 

With the Hamadsha themselves, explanation of the therapeutic proc¬ 
ess is more complex. In the first place, the baraka of the saints enables 
the adepts, the devotees, and the patients to perform the hadra, which 
is pleasing not only to the saints but to the jnun as well. It also enables 
the ghiyyata to play the various rihs which draw the jinn into the 
patient and then expel it. Finally, it enables some of the performers not 
only to enter trance but also to slash their heads; the blood from these 
wounds is again pleasing to the jnun. The baraka of the saints is, then, 
a potentializing force enabling the adepts, the devotees, and patients to 
enter an extraordinary state ( hal) in which they all—in varying degrees 
—have the ability to pass on the baraka of the saints. This baraka is the 
force behind the cure. 

While it is sometimes said that the follower of a saint—an adept— 
possesses baraka which he is able to transfer, just as the descendant of 
a saint can pass on baraka, it is usually claimed that to do this the adept 
must be in an extraordinary state. I have heard this contradicted in 
theoretical discussions, but I have known of only two or three instances 
in which baraka was said to have been transferred to a patient for cura¬ 
tive purposes by a saint’s follower who was not in trance. (In these few 
instances, it was the muqaddim of one of the urban zawiyas—a disinter¬ 
ested party in the transfer of wealth—who passed his baraka to the 
patient.) The entranced state, then, appears to be a key to the cure. 
In this state, the adept may pass on his saint’s baraka by massaging the 
patient, by hitting his head against the patient’s head, by smearing 
blood from his own wounds onto the patient, and by other similar acts 
which are discussed in the next chapter. Often the mere presence of an 
adept in trance is sufficient for the transfer of baraka. Then, too, the 
patient and other devotees who are also in entranced states are said to 
possess the baraka of the saint and may themselves have a curative 
effect. Often they, as well as the adepts, will recite prayers and invo¬ 
cations for participants during pauses in the dance. These invocations 
are frequently recited in an entranced state, and they are considered 
especially effective. 

To summarize, baraka may be transferred either directly or indi- 



168 


THE HAMADSHA 


rectly to the patient. In either case, if the patient is the victim of a 
named-jinn or jinniyya, this baraka is not usually sufficient to effect a 
cure, but rather puts the patient in a state of potential cure. The cure 
is then effected by the patient’s following a regime pleasing to the jinn 
or jinniyya. One important element in this regime is the hadra, which 
permits patient and curers alike to enter trance. Again, the saint’s 
baraka is considered the potentializing force behind the hadra. This 
process is illustrated in Figure 9. 



POTENTIAL FOR HADRA 

* ' 


/ <HA L > V 

SLASHING-Jl DBA- POSSESSION 



ALLEVIATION OF SYMPTOMS 


FIGURE 9. 



9 


The 

Pilgrimage 


There are two principal Hamadsha rituals that are designed, among 
other things, to cure the sick. The first of these, the pilgrimage to the 
tombs of Sidi ‘Ali ben Hamdush and Sidi Ahmed Dghughi, is of more 
general scope and appeal than the second, the performance of the hadra. 
Although it has certain distinctive features, particularly with respect to 
‘Aisha Qandisha, it is not dissimilar to pilgrimages to most of the other 
marabouts in Morocco and elsewhere in the Maghreb. It is by no means 
restricted to adepts or devotees of the hadra, nor is its efficacy limited 
to those suffering from attacks of the jnun. The Hamadsha hadra, too, 
resembles the performances of other popular brotherhoods in Morocco; 
but it, too, differs from them all in certain outstanding respects. Al¬ 
though its efficacy, theoretically, is not restricted to cases involving the 
jnun, its principal therapeutic aim is the cure of the jinn-struck and the 
j inn-possessed. 

The pilgrimage to a saint’s tomb is referred to in Moroccan Arabic 
as a ziyara, 1 a visit, and must not be confused with the hajj , or pil¬ 
grimage to Mecca, which is required of all good Muslims. Although the 
ziyara and the hajj have certain features in common, neither the Ha¬ 
madsha nor other Moroccans of their background seem willing to draw 
any comparison between the two. For the Muslim, there is an absolute 
distinction between a visit to a local sanctuary, no matter how powerful 
the saint involved, and a pilgrimage to Mecca, the seat of Islam. To the 
non-Muslim scholar, however, the structural parallels are of some sig¬ 
nificance. 

1 Ziyara is also used to refer to the offerings one leaves at a saint’s tomb. 



170 


THE HAMADSHA 


In the most general terms, the purpose of a pilgrimage to a saint’s 
tomb is to obtain the baraka of the saint. Often the pilgrim hopes to 
have a dream in which the saint appears and gives instruction as to how 
to deal with his complaints. While visits to saints’ tombs, especially local 
ones, are sometimes made only to receive a saint’s blessing, they are 
usually made with a more explicit purpose. The pilgrim goes to the 
saint’s tomb to ask the saint to intercede with God for him. He is 
anxious to obtain something: a child, a spouse, business success, a favor¬ 
able verdict at court, poetic inspiration, the cure of an illness, the alle¬ 
viation of the symptoms of an attack by a jinn, or simply a change in 
fortune. The saint, who can do no wrong, will intercede for him; the 
success or failure of the mission is dependent upon the not-to-be-ques- 
tioned will of God. Some saints, however, are particularly effective for 
specific things (cf. Voinot 1948). In and around Aleknes, for example, 
Sidi Said is considered especially helpful for coughs and chest diseases; 
Moulay Ahmed Shibli for fevers; Sidi Abdullah 1-Gazzar, the butcher, 
for female alopecia; Sidi Sliman Mula 1-Kifan, the master of the cave, for 
rheumatism and other arthritic pains; and Sidi Abdullah ben Ahmed for 
the insane ( hmeq ). Not all saints are specialists, though, and even 
among those who are, there is considerable variation in their reputa¬ 
tions. It is impossible, strictly speaking, to claim that one saint has more 
baraka than another; but, nonetheless, the saints do appear to fall into 
a hierarchy with respect to their importance and efficacy (Dermen- 
ghem 1954:21). Since the exact relationship between the saint’s inter- 
cessionary power with God and his own baraka is not clear, no one 
can say that the one or the other is responsible for the success of a 
pilgrimage. But while this distinction may be of some theoretical sig¬ 
nificance to the non-Muslim, it is of little concern to the Muslim, since 
all things ultimately derive from God. The pilgrimage to Sidi c Ali’s and 
Sidi Ahmed’s tombs are still more complicated, because they actually 
involve a stop at ‘Ai'sha Qandisha’s grotto. Similar cults are found at 
other mausoleums as well (cf. Brunei 1955:325 et seq.). 

Although a pilgrimage to the mausoleums of Sidi c Ali and Sidi Ahmed 
and to the grotto of c Aisha Qandisha is considered particularly thera¬ 
peutic for barrenness, children’s diseases, and illnesses caused by c ATsha, 
its value is by no means restricted. The pilgrimage is said to be one of 
the most powerful in Morocco, and it is not unusual for visitors, both 
Arab and Berber, to come from as far as Tetouan, Marrakech, or even 
Agadir to obtain some of the saints’ blessing. The majority of pilgrims, 
however, come from the nearby countryside and from Meknes. Al- 




THE PILGRIMAGE 


171 


though pilgrims from Fez are not unusual, Fez seems to fall just outside 
the area of greatest devotion to the Hamadsha saints. 

Pilgrimage Activities 

The pilgrimage to the Hamadsha saints has a number of possible 
stop-offs: the mausoleum of Sidi c Ali and Sidi Ahmed, the c Ayn Kabir 
at Beni Rachid, the grotto to 'Ai’sha Qandisha, and the mausoleums of 
the lesser village saints. It is not incumbent on a pilgrim to make all of 
these stops: he is free to visit any one of the shrines without stopping 
at the others. Once he has made a stop, however, he must follow cer¬ 
tain rules of behavior, which can best be understood as marks of respect 
to the saints and to 'A'isha Qandisha. He is also free to initiate certain 
activities at these stops—activities which, once initiated, are also subject 
to certain rules. The structure of the pilgrimage is not dissimilar to the 
hagiographic legends: there are a number of component themes which 
the narrator is free to recount or not to recount, but once he has 
initiated a theme he must follow it through to the best of his knowl¬ 
edge. The thematic components of the legends do not usually serve a 
straightforward etiological function. They are not usually explanations 
for ritual activities, but they do reflect their relative importance and 
their possible interrelationships. The components of the pilgrimage are 
summarized in Table 4. 

While it is theoretically possible to make only one or two stops, most 
pilgrims, who are anxious to obtain the maximum benefit from their 
trip, visit both mausoleums, the two shrines to c Aisha Qandisha, and the 
c Ayn Kabir. They rarely visit the tombs of the other village saints. The 
pilgrims may spend a few hours or a few days in the two villages. They 
sleep in the courtyard or the tomb room, and bring their own food. 
A trip that lasts a single day is the most common, and one that lasts a 
day and a night is considered ideal. Prolonged visits for weeks and even 
months are very rare at Beni Rachid and Ben Ouarad; they are not un¬ 
common at some sanctuaries, such as Sidi Rahal and Moulay Brahim 
near Marrakech. The Hamadsha shrines are not havens for beggars, 
wandering mendicants, prostitutes, and other social outcasts. No pro¬ 
vision is made for them. 

Decisions to perform optional ritual activities are based on individual 
considerations. The nature of the pilgrim’s complaint plays a role. 
Henna, for example, is commonly smeared on pilgrims suffering from 
aching bones. The most important consideration is economic. Most of 



TABLE 4 

THE RITUAL COMPONENTS OF THE PILGRIMAGE 
Required Conduct Optional Conduct 

Mausoleum 

(Sidi c Ali, Sidi Ahmed , 

Sidi l-Hafycm , other saints) 

Purification Placing head under catafalque 

Removal of shoes Sleeping (and dreaming) in tomb room 

Kissing portals Lighting candle 

Circumambulation of tomb Taking earth 

and kissing four sides of Receiving blessing from muqaddim of 

tomb mausoleum or another descendant of 

Sitting near tomb the saint 

Recitation of fatiha or other Sacrifice (as c ar or gift) 

Koranic verses Sponsoring hadra 

Performing hadra 

c Ayn Kahir 

Drinking water 
Bathing 
Bottling water 

Receiving blessing from muqaddim of 
spring 

Sacrifice (as c ar or gift) 

Sponsoring hadra 
Performing hadra 
Receiving blessing from holy fool 
( 7 najdud) 

Shrines to c Aisha Qandisha 

Sitting (and pressing head against root 
system) in grotto at Beni Raehid 
Tying rag or remnant of clothes to roots 
at grotto 
Leaving amulets 
Taking earth 
Smearing earth on body 
Smearing henna on body 
Burning incense 

Receiving blessing from muqaddim 
Sacrifice (as ‘ar or gift) 

Sponsoring hadra 
Performing hadra 


Purification 
Removal of shoes 
Alms 


Purification 
Alms (or payment) 





THE PILGRIMAGE 


173 


the pilgrims are poor, and most of the optional activities are expensive. 
A blessing from the muqaddim of a shrine or a saint’s descendant costs 
about a dirham; a package of candles 60 francs; a chicken for sacrifice 
from 5 to 10 dirhams; a she-goat about 60 dirhams; a he-goat from 45 
to 75 dirhams; a ewe over 60 dirhams; a ram over 75 dirhams; a cow 
from 150 dirhams; and a bull well over 200 dirhams. The sponsoring of 
the hadra varies from 15 to more than 50 dirhams, depending on the 
instruments used, the amount of time spent, and the performer’s per¬ 
ception of the wealth of the sponsor. (It has to be remembered that 
the average pilgrim spends at least 3 or 4 dirhams for alms, and must 
also pay for his transportation to the Jcbcl Zerhoun.) 

The decision to perform one of these optional activities is dependent, 
then, on the financial situation of the pilgrim and on the nature and 
gravity of his condition. A father whose son is suffering from a para¬ 
lyzed limb will more readily make a sacrifice than a husband whose 
wife complains of rheumatoid pains in her joints. Often, however, the 
decision is not made by the individual pilgrim but by a seer ( talk i 5 ) 
whom he has consulted and whose advice he follows. Then again, the 
performance of one of the options may not be conceived as a matter 
of choice at all, but as a command either by the saint or 'A'isha Qan- 
disha. The command may have been revealed in a dream, or the pilgrim 
may simply have the subjective conviction that it is what the saint or 
c Aisha desires. 

Before making the pilgrimage to the Hamadsha saints—or to any 
saint, for that matter—the pilgrim must purify himself. He is not al¬ 
lowed into the sacred confines ( horm ) of the tomb in a polluted con¬ 
dition. He must bathe carefully after having sexual intercourse. A men¬ 
struating woman is not allowed to enter the mausoleum. Usually, 
pilgrims will go to the baths the day before their trip and will refrain 
from sexual intercourse that night. 


The Pilgrimage Sequence 

The average pilgrim usually arrives at either Beni Rachid or Beni 
Ouarad with several members of his family. If he plans to give a sacri¬ 
fice, he will often bring the animal or animals with him. The pilgrim 
tries to arrive early in the morning in order to spend the entire day at 
the sanctuaries. Although most pilgrims go to Sidi 'Ali’s mausoleum 
first, both because he is considered the more powerful of the two saints 
and because taxis, which are the pilgrims’ most common means of trans¬ 
portation, usually stop at Beni Rachid, those pilgrims who are msannad 



174 


THE HAMADSHA 


to Sidi Ahmed go directly to Beni Ouarad. It is sometimes said that if 
one makes the pilgrimage because of a sick child, one should visit Sidi 
Ahmed first. The average pilgrim brings provisions, and leaves them in 
the courtyard of whichever saint he is visiting. If he has brought a 
sacrificial victim, he gives it to the muqaddim of the sanctuary, who 
may sacrifice it for him during the course of the visit. The muqaddims, 
who are anxious to keep the victim alive to sell, will usually try to per¬ 
suade the pilgrim that giving the animal is sufficient and that there is 
no need to slaughter it immediately. 

The pilgrim removes his shoes before entering the mausoleum. He 
kisses the door jambs, first the right and then the left one, and he may 
drop a few coins into the alms box standing just inside the tomb room. 
Sometimes he waits to give alms until he is about to leave. Women turn 
to the left and walk clockwise around the tomb; men turn to the right 
and make a counterclockwise circumambulation. Both men and women 
kiss the four sides of the catafalque. Once the single circumambulation 
is completed, the pilgrim sits down—women to the left of the entrance, 
men to the right—and mumbles a fatha or a few verses from the Koran 
that he happens to know. At this time the pilgrim is said to experience 
the fear of God ( khushu c ). Occasionally he will experience a trance¬ 
like state which he calls the hal, a term which is also employed for one 
level of trance in the hadra. 

Often, especially if he is sick, the pilgrim will place his head under 
the coverings of the catafalque and press it against the tomb in order 
to have closer contact with the saint and to glean more of his baraka. 
Pilgrims often try to sleep in the mausoleum, hoping to have dreams 
in which the saint appears and gives them instruction. (It is sometimes 
said that the saint is not dead but alive within the sanctuary.) The 
dreams—or, better, the saint’s appearance—are loosely interpreted. 2 All 
males in these dreams, especially when wearing white, are said to be 
saints. Women are often interpreted as 'Aisha Qandisha. If the dream 
is puzzling, the pilgrim may ask the muqaddim of the sanctuary what 
it means. 

The pilgrim may spend only a few minutes in the tomb, or several 
hours. Before leaving, he may light a candle or two, and when he 
leaves he usually takes a little earth with him. This earth, which is called 

'The extent to which these dreams are stereotyped “divine” dreams ( chrema- 
tismoi of ancient Greece), and not simply interpreted in a prescribed manner, 
is impossible to determine. The element of secondary elaboration cannot be ac¬ 
counted for; Dodds (1966:107 et seq.) does not take the latter into due consid¬ 
eration in his work on the Greeks. 



THE PILGRIMAGE 


175 


tuba , 3 has been placed in the sanctuary by the muqaddim and is said to 
contain much baraka. Pilgrims often put it under their pillows to pre¬ 
vent bad dreams or rub it on aching parts of their bodies. Taking the 
earth at Sidi Ahmed’s tomb is said to commemorate Sidi Ahmed’s 
throwing tar and then sand in the eyes of the Sudanese guards who 
pursued him after he had taken ‘A'isha Qandisha and the musical in¬ 
struments from the palace of the King of the Sudan. No legend ex¬ 
plains why one takes a little earth from the tomb of Sidi c Ali. The 
practice is not restricted to the Hamadsha saints and, in fact, is wide¬ 
spread among both Berbers and Arabs. 

The average pilgrim will usually settle down in the courtyard of the 
mausoleum after visiting the tomb. He drinks tea, eats a little food, and 
often gossips with other pilgrims. He may ask the muqaddim for a 
blessing or to arrange for the hadra. The latter is rare, both because the 
hadra is expensive and because most pilgrims prefer to sponsor hadras 
in their own homes, where they can invite friends and fellow devotees 
and thus assure themselves of reciprocal invitations. When he has fin¬ 
ished his meal and is through gossiping, the pilgrim may enter the 
tomb again before moving on to the mausoleum of the other saint. If he 
is at Sidi Ahmed’s tomb, he will usually leave something for ‘Aisha 
Qandisha. He may smear a little of her earth or henna on aching parts 
of his body, or even sacrifice a chicken to her. The pilgrim follows the 
winding path through Qelaa to the second saint’s tomb and follows the 
same procedure there as at the first . 4 

Usually, before leaving , 5 a pilgrim will visit the € Ayn Kabir. There 
he washes his face and hands in the holy water and drinks a little of it. 
Some pilgrims will take a bottle of water home with them. Women es¬ 
pecially are apt to pay the muqaddim of the spring a half dirham in 
order to use the baths. Often the pilgrims receive blessings from a holy 
fool ( majdub ) who has been living in a cave nearby for at least ten 
years . 6 On very rare occasions, a pilgrim will arrange for a sacrifice 
or for the performance of the hadra in front of the c Ayn Kabir. 

The last stop on the pilgrimage is the grotto of ‘Aisha Qandisha in 

* Literally, a clod. This word, which begins with an emphatic “t,” should not be 
confused with tuba , confession or repentance. 

‘Although there are no legends connected with this route, it is generally ac¬ 
cepted as the appropriate path to take. Informants state simply that it is the 
shortest distance between the two mausoleums. They always stress the fact that 
they have taken it. 

“If he spends the night, he sleeps in the mausoleum of the saint he considers the 
most powerful. 

• It is not unusual to find such fools at saints’ tombs; this one is not related to 
the descendants of Sidi c Ali and is not from the village of Beni Rachid. 



176 


THE HAMADSHA 


Beni Rachid. This grotto is visited most frequently by women. Men, 
too, visit it, although many of them consider it a place for women and 
seem embarrassed by their own visits. The male pilgrim will remove his 
shoes, enter the grotto, give something to the muqaddim, and leave. 
Women will spend more time there. They press their heads against the 
exposed roots of the giant fig tree, rub a little earth or henna on their 
bodies, and fumigate themselves with incense (black jawi). The henna 
and incense are brought by pilgrims and prepared by the muqaddim. 
Often the women will tie rags to one of the fig roots as a vow ( c ar) to 
give a sacrifice if their pilgrimage is successful. They may also leave 
amulets prepared by Koranic teachers there. Sacrifices, especially chick¬ 
ens, are not unusual, but the performance of the hadra in the grotto is 
rare. When it is performed, the dancers will often imitate pigs and 
grovel in the mud. 

The pilgrim does not necessarily repeat this cycle of visits if he stays 
at Beni Rachid or Beni Ouarad for several days. Usually he will spend 
most of his remaining time at the tomb of the saint he considers most 
powerful. Often the pilgrim will refrain from sexual intercourse for 
several days after his trip to the Hamadsha saints. He may bring back 
candles, amulets prepared by the saint’s descendants, and bread that is 
also prepared by them. Sometimes he will have to rub himself with or 
drink special herbal brews for several days after the pilgrimage. 

Hadra and Sacrifices 

Both the villages of Beni Rachid and Beni Ouarad, as we have seen, 
have teams of Hamadsha which perform whenever they are asked by 
pilgrims and villagers. Villagers will usually invite a team to their own 
houses, and the performance follows the general lines of the urban cere¬ 
mony. It is held out of doors more often than in the city, especially 
when the ceremony is given not to cure an illness but to commemorate 
an event such as the naming of a child. Ceremonies sponsored by pil¬ 
grims usually take place in the courtyard of the mausoleum. On rare 
occasions they are held in e Aisha Qandisha’s grotto or in front of the 
e Ayn Kabir. I have never witnessed a hadra sponsored by a pilgrim, and 
can only report what I have heard from informants who participated 
in them. Village informants complained that the teams no longer per¬ 
formed as well as they did in the past; the team members are at odds 
with one another and their performances reflect this. Urban and shanty¬ 
town informants who were aware of intra-village factionalism also com¬ 
plained about the ceremonies. They did not find the musicians as gifted 



THE PILGRIMAGE 


177 


as those of the city; the ceremonies were sloppy and not particularly 
satisfying. All that the team members wanted, they said, was money. 
Although I am unable to comment on these criticisms, I believe that the 
pilgrims missed the intimacy of their own ceremonies, in which their 
friends and neighbors participated, and musicians whom they knew well 
performed. To what extent they missed the authority and competence 
of their own muqaddim and ghiyyata remains an open question. 7 

Sacrifices are much more common than the sponsoring of hadras. 
Red, black, or multi-colored (“seven-colored”) chickens are often 
sacrificed to ‘Aisha Qandisha in her grotto. Chickens, as well as goats 
and sheep, both male and female, are also sacrificed to the saints. A ram 
is considered the “best” sacrifice. Ideally, the muqaddim slits the throat 
of the animal in front of the saint’s tomb or in ‘Ai'sha’s grotto, follow¬ 
ing the procedure employed in the slaughtering of all animals. The 
meat and skin are then divided among the descendants of the saint, 
and a little of the meat may be given to the pilgrims. In fact, the muqad¬ 
dim usually tries to dissuade the pilgrim from having his animal killed. 
Often, if the muqaddim succeeds, the animal is then sold to pilgrims 
who want to give a sacrifice but have not brought one to the village 
with them. Through a continued process of resale, the children of the 
saint are able to maximize their profits. 8 

Such a sacrifice is referred to as a debiha , or as an c ar, and must not 
be confused with the dhiya, the sacrifice of a sheep on the c Ayd 1-Kabir 
(The Feast of the Sheep), which commemorates the Abraham and 
Ismail story. The dhiya, a yearly duty of all good Muslims, was recog¬ 
nized by all my informants as qualitatively different from sacrifices to 
the saints and to ‘Aisha Qandisha. The element of commemoration is 
not present in the latter: they are conceived, as Westermarck (1926(1): 
188 et seq.) points out, as either an c ar or a gift ( hadiyya ). A debiha is 
considered an c ar when the pilgrim has come with a request, and as a 
gift when he comes to thank the saint or ‘Aisha Qandisha for granting 
a request made earlier. The gift-sacrifice, in fact, generally results from 
a vow ( *ar) to give an animal if a request is granted. This vow may be 
stared in public, or it may simply be a promise to oneself to give the 
sacrifice. The pilgrim may symbolically affirm his vow by tying a cloth 
to the roots of the giant fig tree in ‘A'isha’s grotto or to the branches of 
the fig tree in the urban lodges. Sometimes a first sacrifice is offered, 
with the promise of another when the request has been fulfilled. 


7 A detailed description of the hadra will be given in Chapter Ten. 

"A similar procedure is followed at Aloulay Idriss, even at the musem. 



178 


THE HAMADSHA 


M., who had been struck paralyzed by 'Ai’sha Qandisha, came to Sidi c Ali’s 
mausoleum with a ram and had it slaughtered in front of the saint’s tomb. 
He referred to this sacrifice as an *ar to the saint. He explained that he had 
placed himself under the protection of the saint and had asked the saint to 
intervene with God and c Aisha Qandisha on his part. He vowed that if he 
were cured, he would make another sacrifice of a ram to the saint. Several 
days later—he was still at Sidi c Ali’s tomb—his condition improved after 
dreaming that the saint released him (ordered him to leave). He bought a 
ram from one of the saint’s descendants—this ram had been brought by 
another pilgrim as a sacrifice, but had never been killed—and had it sacri¬ 
ficed in front of the tomb as he had promised. This time, he referred to the 
sacrifice not as an 'ar but as a debiha, and explained that it had been made 
in fulfillment of his first ‘ar. He added that were he not to make it, he 
would incur the anger of the saint and c A'isha Qandisha and would fall sick 
again. 

While it is possible to distinguish analytically between the sacrifice as 
an c ar and as a hadiyya, it is my impression that in the minds of most 
pilgrims who offer a sacrifice this distinction is not so neat. The e ar is 
also considered a hadiyya. 


Fatima’s Pilgrimage 

What follows is a verbatim account of a pilgrimage to the Hamadsha 
saints. Fatima’s mother, Fatna, told the story. Fatna, who lives in Sidi 
Baba with a son and four daughters, is by bidonville standards a com¬ 
paratively wealthy woman. Her husband spends most of the year in 
France as a worker for a construction company. She is not a Ham- 
dushiyya, but she considers herself to be supported (msannad) by Sidi 
c Ali and has had several of her other children’s hair cut in the mauso¬ 
leum of the saint. She attends the annual pilgrimage at Sidi c Ali when¬ 
ever she can and often watches the Hamadsha’s public performances in 
front of Sheikh al-Kamal in Meknes. Although she does not consider 
herself to be dependent ( mwwala) on 'Aisha Qandisha, she is afraid of 
the she-demon and is worried about becoming ‘Aisha’s victim. Never¬ 
theless, she is emphatic about not being a devotee of the Hamadsha, 
and she does not want her children to become muhibbin.® 

I will tell you about the time I took Fatima to Sidi c Ali. Fatima was 3 years 
old then. [Further calculation revealed that Fatima was actually not quite 2.] 
It was at harvest time. My husband was back from France. The first sign 


8 The account was told to my wife. 



THE PILGRIMAGE 


179 


of Fatima’s illness was her falling down on die spot. She was feverish. She 
nearly fainted. Her eyes and her mouth were wide open. She cried and 
cried. She kept crying. I was pregnant with Najiya—not the Najiya you 
know [one of Fatna’s daughters], but with the Najiya who died. I was 
about six months pregnant. [It was established that Fatna had not nursed 
Fatima for three months when the girl fell sick. Moroccan women stop 
nursing an infant the moment they know they are pregnant again. They 
believe the milk of a pregnant mother is poisonous. It is said that it is 
rendered poisonous by the embryo, who is jealous.] Fatima was sick at 
night. I touched her, and she was burning. She was sleeping next to me 
[along with Fama’s husband and two other children]. The moment she 
cried, I took her on my knee and then on my back and then again on my 
knee. She continued to cry a lot. I gave her milk, but she refused it. I gave 
her something to eat, but she refused that too. Then I knew that Fatima 
must be very sick. . . . 

In the morning I decided to go to the fqih [the Koranic teacher]. My 
husband said, “No, it is necessary to go to the hospital.” I answered, “No, 
it is not an illness for the hospital. It is an illness for the fqih.” Then my 
husband let me go. [Fatna’s husband’s family lived next door and en¬ 
couraged her to go to the fqih and then, if the fqih was unable to help, to 
go to a saint.] I, too, knew that it was the illness for the fqih and the 
saint. When an infant falls sick and has the fever and breathes with diffi¬ 
culty, it is an illness which is for the fqih. But when a child eats and 
vomits, when he is sick in the intestines, then it is an illness for the hospital. 
Also if he coughs [chokes] when he vomits, it is an illness for the hospital. 
I learned to tell them apart from the neighbors. . . . 

So, that morning, I went to the fqih. It costs 100 francs. I had not visited 
that fqih before. One of my neighbors, a woman, had told me to go to that 
fqih because he is especially good for children. I went to the house of the 
fqih with the neighbor who recommended him. I said to him, “My daughter 
is sick. Can you write an amulet for her?” The fqih looked at Fatima who 
was feverish, but he did not examine her. [The latter observation was in 
response to questions previously asked Fatna about fqihs.] “It is all right. 
She will not die.” Then he wrote a verse from the Koran. He told me to 
put the verse along with some herbs—rue [/*j77], harmal [kormal"], and a 
type of absinthe [shiba ]—in a red cloth and hang it from Fatima’s neck. He 
also gave me another verse of the Koran and told me to put it in a bowl with 
some water, olive oil, and garlic. I was to rub the mixture all over Fatima. 
All over her, except the soles of her feet and the palms of her hands. I was 
to do this every morning and every night, twice a day, for three days. I 
was also to give Fatima a little of it to drink with each application. [Did 
the fqih ask you anything?] The fqih said absolutely nothing else. You go 
to him, really, only to know if the child will live or die. Three days later, 
Fatima was still sick. 

My mother-in-law said, “Now you must try mint. You pound the mint into 



180 


THE HAMADSHA 


a powder and moisten it with water. You put it in a rag and squeeze out 
drops into the ears, mouth, and nostrils of the child.” I did this to Fatima. 
During the three days I followed the fqih’s instructions, my husband kept 
telling me to go to the hospital. My mother-in-law said that we must try 
mint. After we had tried it, Fatima was a little better. She was still sick, 
though. 

Now before having Fatima, a neighbor—she was a midwife—dreamed that 
I had washed a sheepskin rug in the fountain. The hair of the sheepskin had 
been shaved, and it was smooth. After she had the dream, the midwife came 
to my house and said, “When you have your child—be it a boy or a girl 
—you must take it to Sidi 'Ali to have its head shaved by one of the chil¬ 
dren of the saint.” I was pregnant then with Fatima, and after she was born, 
when she was forty days old, I took her to Sidi c Ali, and the muqaddim 
[of the mausoleum] shaved her head for 300 francs. Afterward, I said to 
him, “If all goes well with my child, I will return in a week’s time with a 
chicken for you.” 10 Then I went home. 

After a week I saw that Fatima was well, and I thought that it was 
not worth going back to Sidi c Ali, not worth spending a lot of money 
for a chicken. So I never went to him with the chicken. Now when my 
child took sick, I had this point—not having given the chicken—between 
my eyes. The muqaddim had shaved Fatima’s head in the saint’s mausoleum, 
right next to the tomb, and I had not returned with the chicken I had 
promised him. When Fatima fell sick, my husband said to me—and all the 
neighbors said to me—that it was because of the chicken. Two days after 
I tried the mint, I went to Sidi 'Ali. 

There is one more thing. On the day I gave Fatima mint, I went to Moulay 
Abdullah ben Ahmed [a saint who is buried a mile from Fatna’s house]. 
I left the verse of the Koran the fqih had given me there. Moulay Abdullah 
ben Ahmed is where one takes people who are crazy. I was walking by 
with Fatima on my back. I was on my way to the madina to shop when I 
passed the mausoleum and decided to leave the verses there. When a fqih 
writes verses of the Koran, and they do not work, we leave them in a 
saint’s tomb. If the verses work, you must always keep them. But when 
they do not work, you leave them in a saint’s mausoleum because you can¬ 
not just throw them anywhere. Perhaps someone will walk on them and 
dirty them. There is always a shelf made of earth in a saint’s tomb where 
you leave them. They are said to be thrown away and will not work again, 
not even from the saint’s tomb. 

Now then, everybody was saying, “It is because of the chicken that Fatima 
is sick.” This was always between my eyes. Everybody said, “Go back to 
Sidi c Ali so that God can help you.” The woman who had taken me to 
the fqih said, “Go to Sidi e Ali.” And although Fatima was a little better, 
my mother-in-law said, “I know that unless you give a chicken to the saint, 


10 Fatna does not mention that her child was sick at this time. 



THE PILGRIMAGE 


181 


Fatima’s sickness will last forever.” The day after I used mint, I said to my 
husband that I was going. I did not do anything for Fatima that day. My 
husband bought three chickens. He agreed to my going, but he said, “You 
must find a neighbor woman to go with you. I will pay for her trip.” 
[Fatna’s husband, having worked in France, seems to have considered him¬ 
self too modem to go with Fatna.] I tried my neighbors, but they were all 
busy. So my husband had to come. He bought the chickens the day before 
the trip. He ordered a taxi. For Sidi Ahmed, he bought a black and white 
chicken; for Sidi c Ali, a red and white chicken; and for 'ATsha, a chicken 
that was all red. Also, on the day before the trip, my husband went to the 
baths. I could not go because Fatima was very sick. I did not want to leave 
her. 1 washed very well and heated water for it at home. I washed well 
because it is not good to go to a saint when you are dirty. My husband 
bought a box of candles. There were 24 in the box. We also took some¬ 
thing with which we could cut off a piece of Fatima’s clothing to hang on 
c ATsha’s tree. You say that on that piece of clothing, you have left the ill¬ 
ness. [Did you wear any special clothes?] We put on our best clothing. 
We all had on our newest clothing except for our jallabas. I put new 
clothes on Fatima too. You should wear good clothes to the saint. They 
should be the best clothes, because it is possible you will meet people you 
know at the saint. From the point of view of the saint, you can wear new 
or old clothes. The saint can’t see you. You must be clean. ... I put on 
my new clothes for the trip. [Do you have to wear special colors?] It is 
not necessary to wear any special colors when you visit a saint. If a child is 
sick, you must not wear any perfume. You can wear other make-up. You 
bring whatever verses of the Koran the fqih has written for the child, and 
you leave them in 'Aisha’s grotto. [Fatna explained that they were attached 
to the roots of the fig tree.] Whatever words of the Koran you are wearing 
when you make a trip to Sidi 'Ali, you leave with c AIsha—even those which 
work. [This practice is by no means universal.] So, when we went to Sidi 
'Ali, my husband wore European clothes. It makes no difference for men, 
European or Moroccan clothes. Women should wear Moroccan clothes. 
It is more respectful. My husband had ordered the taxi for 8 in the morn¬ 
ing. We were not planning to stay the night, because we wanted to keep 
the same taxi and return by noon, in time for lunch. The taxi driver had 
arranged to make the trip for us for 2,000 francs. 

The taxi went directly to Sidi Ahmed. The moment we got into the taxi 
Fatima was better. At Sidi Ahmed, I put Fatima under the covers [of the 
catafalque]. My husband came in with me but left immediately because he 
was bored and there were too many people there. Before leaving, he gave 
the chicken to the muqaddim and left eight candles there on the alms box. 
He did not leave any money. When you enter a mausoleum, you kiss both 
sides of the door and the tomb. Fatima was too young to do that. But at 
Sidi Ahmed’s and then at Sidi 'Ali’s I rubbed her head and her face against 
the tomb. I left Fatima under the covers for about 30 minutes. She did not 



182 


THE HAMADSHA 


cry. She had been getting better since we got into the taxi. After 30 min¬ 
utes, I took Fatima into the courtyard to Lalla ‘Ai’sha’s hole. I sat down by 
the hole and took some earth [tuba] and rubbed it on Fatima’s chest, 
forehead, and arms. It was Friday, if I remember. You visit the saints on 
Friday or Thursday. It is best then. 

Then we went quickly to Sidi e Ali, because the taxi was waiting. We fol¬ 
lowed the footpath, but the taxi went by the road. At Sidi ‘Ali’s we entered 
the tomb, kissed the jambs and the tomb. I rubbed Fatima’s hands and face 
on the tomb and put her under the covers. I kept her there for 45 minutes. 
My husband stayed in the tomb room. He gave the chicken to the muqad- 
dim and left eight candles. Fatima was not crying and was no longer fever¬ 
ish. I talked with my husband. He said, “You see, had you taken her here 
on the first day, she would have been better. It is all your fault because of 
the chicken.” I had put Fatima under the cover, because one says that in 
that way you make an ‘ar with Sidi e Ali in order for the child to get better. 
The child is given to the saint for his protection. The child is like a sheep 
given to a saint [for sacrifice]. After you have made an c ar with Sidi ‘All, 
the child belongs to Sidi e Ali. If the child gets better, he will get better 
right away. If he does not get better, he will die immediately after the visit. 
Sometimes the child will die right under the covers. So, like the sheep, you 
say, “This child, O Sidi c Ali, belongs to you.” We left no money at Sidi 
e Ali, but we left the chicken and the candles. 

Then we went to 'Alsha’s grotto. We gave the chicken to the muqaddim, 
as well as the last eight candles. Then we cut a piece of Fatima’s clothing 
and hung it on the tree. I had a verse of the Koran hanging from my neck. 
I left that, too, at Lalla ‘Aisha’s. My husband came with me to the grotto. 
I took some henna and put it on my forehead and on my chest and also on 
Fatima’s forehead and chest. You find this henna in ‘Aisha’s grotto. We left 
right away. There was no place to sit. We were the only people there. It 
was frightening, with the spring, with the tree. It is always dark there, even 
during the day. [Do you kiss anything there?] You do not kiss anything 
there. 

Fatima was completely cured by the time we got home at noon. We 
did nothing for Fatima afterward. We left her in the clothes she had worn 
for three days without washing her. We kept her in the same pants 
too. We took them off when we knew she had to poopoo. Then we would 
put them on again. When she made peepee, we took them off to dry, but 
we did not wash them. This we did for three days. I changed my clothes 
immediately. So did my husband. My sister-in-law made us lunch, and Fatima 
ate for the first time since she had fallen sick. She had eaten nothing for 
five days—just a little water. We say that hunger docs not kill until after 
40 days. After lunch we slept. We slept and slept until 4 o’clock. Then we 
had a simple dinner of rice and milk. Before dinner, at 5 o’clock, all the 
friends and neighbors came over to congratulate us. They stayed until din- 



THE PILGRIMAGE 


183 


ner. They drank tea with us. After dinner, we, the family, all talked until 
10:30.” 

Ah, there was one thing 1 forgot. After we returned, there were two things 
to do: one for Sidi Ahmed and one for Sidi ‘Ali. For Sidi ‘Ali, you take a 
little mint and squeeze it into the cars, nostrils, and mouth of the child. 
We say that then there will always be flowers in these places. For Sidi 
Ahmed, you take a little of the earth from the hole in the courtyard—the 
muqaddim gives it to you—and you put it in a rag and hang it from the 
child’s neck. I did this for Fatima. I also prepared the mint for her. It is 
important not to forget the mint and the earth after a visit to Sidi ‘Ali and 
Sidi Ahmed. Also, after a visit to Sidi Ahmed, the mother of a sick child 
must not eat butter for seven days. She must not put henna or perfume on 
herself. So I did not eat butter for seven days, or put on henna, or per¬ 
fume myself. That is all I did. 

I have presented Fatna’s account of her pilgrimage in detail because 
it illustrates what one does on a pilgrimage and, more important here, 
how one makes the decision to pay a visit to the saints . 12 Such decisions 
are clearly not made on the basis of the objective condition of the 
patient alone. Although the patient’s condition does delimit the range 
of choice, it does not specifically determine a particular cure. Factors 
not directly involving the patient play an important role. In this case, 
Fatna’s decision to go to Sidi ‘Ali—even when Fatima’s condition was 
beginning to improve—seems to have been based on a conviction that 
she was responsible for her daughter’s illness. She had neglected to give 
the chicken, as she had promised, to the muqaddim of the mausoleum. 
She either was encouraged in this belief by her neighbors or projected 
her own conviction onto them. 

The Westerner would argue that Fatna had a guilty conscience and 
was determined to expiate her sin of omission by means of a pilgrimage 
and the sacrifice. Moroccans of Fatna’s background, do not, however, 
organize their experiences in terms of guilt, sin, and expiation. They 
organize experience in a more externalized way, and would argue that 
Fatna’s failure to give the chicken to the muqaddim angered either Sidi 
‘Ali or ‘Ai'sha Qandisha and thus resulted in her daughter’s illness. Fatna 
does not specify who was angry, but we may assume that, since saints 
do no harm, it must have been ‘Aisha who was responsible for Fatima’s 

n The information in the above paragraph was in response to a number of 
questions put to Fatna. Her story of the pilgrimage ended with the family’s 
return to Sidi Baba at noon. The final paragraph of her narration was again her 
own. 

u Cf. Crapanzano and Kramer 1969 for an account of another pilgrimage. 



184 


THE HAM ADS HA 


condition. This, at any rate, is how similar cases were explained to me. 
The pilgrimage and the sacrifice of the chickens—they were not ac¬ 
tually slaughtered—were means of appeasing the saints and £ AYsha 
Qandisha. Fatna placed Fatima under the protection of Sidi £ Ali (and 
Sidi Ahmed)—“The child, O Sidi e Ali, belongs to you”—and, in so 
doing, abrogated her own responsibility. The child’s fate depended 
then upon the saint. “If the child gets better, she will get better right 
away. If she does not get better, she will die immediately after the 
visit.” She does not specify whether the child was also placed under 
£ AYsha Qandisha’s protection. 

The role of £ Aisha Qandisha in these matters is very ambiguous. 
Although sacrifices to £ AYsha are often referred to as £ ars, it is my im¬ 
pression that they are generally conceived of as gifts to placate the she- 
demon. £ AYsha is said to like red or black chickens or to like blood. 
One does not give a child to her. It must be remembered that while it 
is possible for a human being to have a child msannad by a saint, it 
is not possible for a human being to make a child a follower of a jinn. 
It is the jinn who decides. Since both saints—especially Sidi £ Ali—have 
some influence over £ AIsha Qandisha, one places the child in their hands 
and hopes, Allah willing, that they will exercise that influence and that 
c AYsha will accept it. The sacrifice to c Aisha is all that the pilgrim 
himself can do to placate her. 

It has to be remembered that while the child’s illness is the raison 
d'etre for the pilgrimage, the pilgrimage serves primarily to remedy a 
situation which affected the child, but for which the child was not at 
all responsible. Just as Thebes suffered the plague because of Oedipus’ 
conduct, so Fatima suffers because of Fatna’s negligence. Therapy is 
not patient-centered but situation-centered. 



10 


The 

Hadra 


A ceremony at which the hadra is performed is most commonly 
called a lila y or night, because it usually is held at night. When it is 
held in the morning or the afternoon, it is called a taqyil; and in the 
late afternoon and early evening an * ashwiyya . Whenever the cere¬ 
monies do not involve the recitation of litanies, they may be referred 
to simply as hadras. Fraja, which refers to a secular spectacle, is very 
occasionally used to refer to the public performances of the hadra. It 
is never used for private ceremonies. 

One very common name for the Hamadsha ceremonies—and for 
those of other brotherhoods, especially when they are terminated by 
a communal meal—is sadaqa. “Sadaqa” means charity, gift, or alms, 
and is frequently used for the alms left by a pilgrim at a saint’s tomb. 
It is also used to refer to money given the Hamadsha by spectators 
and devotees, and the money and gifts that the adepts of the urban 
lodges divide among themselves. In this sense, baraka may be substi¬ 
tuted for sadaqa. One of its derivatives, sdaq, refers to the bride-price 
that a husband must pay to his wife’s family. Mercier (1951:191) sug¬ 
gests that “sadaqa” may also mean the expiation of sins, but one has 
to be extremely cautious in dealing with the implications of such a 
definition. The adjectival form sadiq means honest, loyal, truthful, or 
sincere. A man who is sadiq is a man who lives up to the highest 
standards of the Islamic community. Its verbal form, sadaq , means to 
tell the truth or to act with sincerity, and the Hamadsha emphasize 
that their ceremonies must be performed with sincerity by men who 
are sadiqin. If a man is not sincere, not only will he not receive the 
blessings, the baraka, of the Hamadsha saints, but he wdll suffer the 
dire consequences of c Ai’sha Qandisha’s wrath. 



186 


THE HAMADSHA 


The ceremonies of the urban lodges are much more complicated 
than those of the bidonville teams. They usually involve the chanting 
of litanies as well as the performance of the hadra, whereas a team cere¬ 
mony consists only of the hadra. Occasionally a few lines of a litany 
( dikr ) will be chanted at a team ceremony, if the members happen to 
know any, but chanting is not central to the ceremony and it is gen¬ 
erally ignored by the devotees, who are anxious to get on with the 
hadra and dance. The adepts of the urban lodges are called in to chant 
their litanies alone; these invitations are made on festive occasions, 
such as name-days or moving days, and they are said to bring much 
baraka to the sponsor and his family. The zawiya members never per¬ 
form in public as their confreres in the bidonvilles do. I have seen 
them in attendance at public ceremonies, but never actively partici¬ 
pating. Their ceremonies are less violent, less spectacular, than those 
of the shantytown Hamadsha. 

The adepts claim to work—to perform their ceremonies—out of 
love for the saint they follow. They are much more explicit in their 
devotion to the Hamadsha saints than are the shantytown Hamadsha, 
who are more concerned with c ATsha Qandisha and the other jnun. 
The adepts conceive of their performances as a means of obtaining 
their saint’s baraka—baraka which keeps them in good health and 
fortune, and which they can pass on to others during their ceremonies. 
It is by virtue of their saint’s baraka that they are able to perform the 
hadra and fall into trance ( hal ). Both the urban and shantytown 
Hamadsha would consider any suggestion that their hal is a mystical 
union or communion with God, or their saint, as blasphemous. The 
hal is an end in itself, a revitalizing process. I have been told that 
awakening from the trance is like being reborn. The Hamadsha per¬ 
form the hadra because their saints did so before them. The saints’ 
performances were pleasing to God, who gave them their baraka, and 
now they, the Hamadsha, their followers, obtain their own baraka 
from them. 

Since the bidonville Hamadsha are more concerned with the jnun 
than with the saints, they perform the hadra to please c Aisha Qan¬ 
disha and the other demons. It is the saints who have given them the 
power to perform the hadra, but it is c Aisha Qandisha who enables 
them to trance and forces them to slash their heads. Sidi c Ali ordered 
Sidi Ahmed to fetch the hal, and this he did by bringing c A'isha 
Qandisha from the Sudan. The zawiya members, too, acknowledge 
the importance of c Aisha Qandisha, but they are more reticent in 
talking about her and usually prefer to emphasize the role of Sidi e Ali 
and Sidi Ahmed. 



THE HADRA 


187 


Both the madina and shantytown Hamadsha recognize the efficacy 
of their performances in the cure of illnesses, particularly those ill¬ 
nesses caused by 'Aisha Qandisha and other jnun. The madina adepts 
perhaps lay less stress on the therapeutic effect of their saints’ baraka 
than do the team members from the shantytowns. They conceive of 
their ceremonies as a means of passing baraka to everybody, both 
the sick and the healthy. The team members see their ceremonies as 
pleasing to the jnun—“ 'Aisha likes a particular rih,” “ 'Aisha likes the 
sight of blood”—and thereby therapeutically effective. Their cere¬ 
monies arc directed to the sick primarily. 

The distinction between the two groups is difficult to communi¬ 
cate. It is one of tone and emphasis and not of belief. Belief itself is 
ambiguous and subject to many interpretations. 1 The adepts of the 
zawiyas are treated by both the Hamadsha and other Moroccans with 
great respect; they are the poor, the foqra, the sadiqin, who strive to 
live up to the ideals of the Islamic community as they themselves 
understand them. They are men possessed of baraka, and their cere¬ 
monies must be respected regardless of whether they are viewed with 
approval or disapproval. The team members of the bidonvilles are not 
treated with this same respect; they are often referred to as charlatans, 
money-grubbers, and whoremongers. Their ceremonies are accepted 
as necessary; they may be viewed with wonder and with awe, but 
never with respect. This difference is reflected in my own reactions 
and the reactions of my informants to the ceremonies of the two 
groups. I have been told that at the zawiya ceremonies one has the fear 
of God; at the bidonville ceremonies, the fear of 'Aisha. Both my wife 
and I recognized a harsh and often histrionic quality in the bidonville 
ceremonies—a violent theatricality and an ambience of threat and 
even terror—which we never felt at ceremonies of the madina. Our 
reactions were confirmed by several of our non-Hamadsha friends. 


The Madina Ceremony: Preparations 

Once the decision to sponsor a ceremony has been made, the sponsor 
tells the muqaddim of the zawiya he prefers. 2 Date and time are set— 
most ceremonies begin between 7 and 8 in the evening—the fees for 

1 The ideology of the < Allaliyyin, under the leadership of their new muqaddim, 
appears to be changing in the direction of that of the more active bidonville 
teams. The Dghughiyyin have maintained a more conservative position. 

* In what follows I assume for clarity’s sake that the sponsor or host is not the 
patient, but a relative. Usually he is the father or elder brother of the patient, or 
the husband in the case of a married female patient. 



188 


THE HAMADSHA 


the ghiyyata are discussed, and a deposit of 5 or 10 dirhams is left. 
The deposit, or at least 5 dirhams of the deposit, is given to the khalifa 
of the zawiya, who must let all the adepts know of the performance 
and bring the ritual paraphernalia to the sponsor’s house on the ap¬ 
pointed day. The paraphernalia consists of the zawiya’s flag, a basket 
for collecting alms, and a few spare drums ( gumlat ). The muqaddim 
may arrange to invite a few of the zawiya’s devotees, especially ones 
who are in need of a ceremony, but most of the guests are invited 
by the sponsor and his wife. They include members of the family, 
friends and neighbors, and any devotees the sponsor happens to know. 
Theoretically, anyone else who hears about the ceremony is welcome 
to attend. In fact, only acquaintances of the host and devotees who are 
driven to enter by the sound of their particular musical phrase attend 
the ceremony uninvited. Usually between 45 and 80 people will come. 
Some will stay for the full performance, others for just a few minutes. 
Only invited guests will remain for the meal with which the ceremony 
concludes. 

The host, his family, and the patient will usually go to the public 
bath on the day of the ceremony. Others will bathe carefully at home 
before attending the hadra, if they do not go to the baths themselves. 
While technically no one should perform the hadra in an unclean con¬ 
dition, the fact that participation is not altogether voluntary means 
that great emphasis cannot be placed on ritual cleanliness. A man, clean 
or unclean, who hears his rih cannot resist dancing without incurring 
‘AYsha’s wrath or the displeasure of his own particular jinn. This, of 
course, has dire consequences. 

If there is going to be a sacrifice, it is usually performed by the host 
the morning before the ceremony in the presence of his family, the 
patient, and perhaps a few friends. The muqaddim or the khalifa of 
the zawiya may sometimes help him. 3 A ram is the ideal sacrifice, but 
often a male or female goat, which is cheaper, is slaughtered. Although 
the sacrifice is not usually central to a madina ceremony, it is of con¬ 
siderable importance to the host, since it is his major expense. Goats 
may cost as little as 45 dirhams, and rams as much as 200 dirhams or 
more. The meat of the animal is used to make the couscous and stew 

* Sometimes the muqaddim will force the patient to collect the blood of the 
sacrificial animal in his cupped hands and swallow it. He may smear a little blood 
from the knife on the patient’s forehead and neck. This practice is very rare 
among the madina Hamadsha but, despite Koranic interdiction (Sura II, 1, 73), 
it is not infrequent among the bidonville and rural Hamadsha. When it is fol¬ 
lowed, the sacrifice takes place at the beginning of the formal ceremony after 
the guests have arrived. 



THE HADRA 


189 


( tajin ) that are served at the ceremony to the adepts and the guests. 
The skin is given to the zawiya. Goatskins are used for drumheads; 
sheepskins may either be given to the saint’s descendants or sold and 
the proceeds given. If no sacrifice is made, meat is bought from the 
local butcher. After making the sacrifice, the host will spend the rest 
of the day waiting for guests and entertaining early arrivals. Some¬ 
times, these early arrivals are members of his family who have come 
from far away to attend the ceremony. The host’s wife, usually aided 
by neighbors and girls and women in the family, spends the day pre¬ 
paring the ceremonial meal. 

That evening, the adepts and ghiyyata parade in front of the host’s 
house to attract attention. The host usually stands at his door and 
greets them. They play a few of the popular rihs and dance a little. 
There is seldom any trancing at this stage. They will be surrounded by 
excited men, women, and children, who often follow them into the 
sponsor’s house. Once inside, the muqaddim repeats a fatha 4 for the 
host and his family. A special prayer is usually said for the patient. 
The muqaddim or a sharif may massage or spit on the ailing part of 
the patient’s body. Then all of the adepts sit down in a comer of the 
courtyard, and the host tends to them and the other guests. Men remain 
in the courtyard when the weather permits it; women stay in a sepa¬ 
rate room or on the roof. Both the men and women sit or stand pressed 
close together, as they do on all ceremonial occasions. While the 
guests wait for the host to serve them milk and dates—a sign of wel¬ 
come and early return—and mint tea, the indispensable ingredient of 


4 Fatha is the dialect form of fatiha , which refers to the opening verses of the 
Koran. Fatha is used bv the Hamadsha and other Moroccans of their background 
for an invocation which does not necessarily contain the Koranic verses. A typical 
fatha at a f^amadsha ceremony is: 

O my brothers, pray for me. 

I shall be cured, thanks to your invocation. 

Shelter me under your wings. 

O children of Mustafa, the chosen, the elected. 

There is no God but Allah. 

O Prophet, cure me. 

All of the lines but the last are repeated three times. Sometimes the fatha is much 
shorter. The following one was repeated at a curing ceremony in a bidonvillc. 


Thanks to Allah and Sidi c Ali. 

Allah will give you bread that is not 
prepared by the hands of men but by angels. 


This fatha, repeated by a sharif, was considered to be particularly good. One of 
the listeners described his reaction to it in the following words: “I felt my heart 
to be very hot at this time and then I heard the fatha and then it was as though 
a pail of cold water—cold wind—made my heart go cold.” 



190 


THE HAMADSHA 


Moroccan hospitality, the adepts gossip among themselves. The drum¬ 
mers may tighten the membranes of their drums over a brasier at this 
time. 


Litanies 

When all of the guests are served, the muqaddim leads the adepts 
in a responsive recitation of the hizb: a long, laudatory invocation to 
Allah, whose forgiveness is sought. 

A. In the name of Allah, the compassionate and the merciful 

B. May eternal happiness rest with our lord Muhammad, His family, 
and His companions 

A. Let us praise Allah the benevolent 

B. Him, the adored Allah 

A. May He hear my wishes 

B. May He pardon me and grant me His mercy 

A. I am a sinful slave 

B. And I implore my lord 

A. Because it is my duty 

B. To ask Him to guide me 

A. O, the compassionate one, the merciful one 

B. Protect us from hell 

A. You are the only one, the unique 

B. You are the unique and the omnipotent one 

etc. etc. 

This is a section of the hizb recited by the Dghughiyyin of Meknes. 
The 'Allaliyyin zawiya has a different hizb. There is in fact no stand¬ 
ard hizb for either Hamadsha brotherhood, and differing texts are 
used throughout the country. It is claimed that the hizb of the ‘Alla¬ 
liyyin of Meknes was composed by al-Shadhili, but this seems unlikely, 
since al-Shadhili left no writings (Rahman 1968:197). The Dghu¬ 
ghiyyin do not know the origin of their hizb, but I have been told 
by a Fasi in-law of one of Sidi ‘Ali’s descendants—a man whose wisdom 
is generally respected—that it was composed by Sidi Heddi, the 
founder of the mendicant brotherhood known as the Heddawa. The 
language of the hizb and of the other litanies is written in what the 
Hamadsha refer to as classical Arabic but which is, in fact, a mixture 
of classical and dialectical Arabic. Most of the adepts do not under¬ 
stand it very well and are content reciting it. 

The muqaddim takes no special place when he leads the recitation. 
He remains seated with the adepts, who are pressed together and who 




THE HAORA 


191 


may sway slightly from side to side. The adepts do not engage in any 
of the gymnastics reported for other brotherhoods; they often close 
their eyes during the recitation of the hizb. From time to time, one 
adept or another will replace the muqaddim as leader of the recitation. 
The men in the audience center their attention on the adepts, but they 
may occasionally gossip or finish drinking their tea. The women seem 
less interested in the recitation of the hizb and other litanies. The 
patient—or patients—is generally ignored. 

Sometimes, in place of the hizb, short moralistic tales and prayers 
called ivanasa are recited. The most popular of these songs are said 
to have been composed by al-Maghrawi, a bard in the court of the 
Saadian sultan al-Mansur ad-Dhabi (1578-1602) who was considered 
one of the greatest of the composers of dialectical lyrics ( malhun) 
(Chottin 1938:104). A pilgrim, for example—I am not sure whether 
this poem is by al-Maghrawi—makes no provisions for his journey to 
Mecca, but he is aided by another “pilgrim” and has a successful 
journey. 


I made a pilgrimage with faith and conviction. 

I came back happy and with changed name. 

All listeners called me hajj. 

My heart was filled with joy. 

Allah satisfied my wishes. 

I came back surrounded by friends and brothers. . . . 

When the hizb or the wanasa is completed—it lasts anywhere from 
about five minutes to over an hour—there may be a pause for tea 
before the recitation of the dikr begins. The dikr, resembling the hizb 
in content, consists of shorter phrases which are said more rapidly. 
The adepts may accompany the recitation with hand-clapping and a 
gentle tapping of drums. The ganbri may be played. 

A., B. In the name of Allah, the compassionate and the merciful 

A. I begin with the name of Allah 

B. To him I dedicate the following words 

A. Muhammad, O Perfect Creature 

B. You are as soft as beeswax 

A. You, who have brought the light 

B. You, who have come as an envoy 

A. You are the enlightened man 

B. You are the savior at the last judgment 

etc., etc. 
(Section of dikr used 
by Dghughiyyin) 



192 


THE HAMADSHA 


Although the adepts are still seated, they do sway back and forth 
more rapidly as they recite the short dikr phrases, and they also start 
to hyperventilate. I have been told that they have the fear of God at 
this time. As the dikr is repeated faster and faster and louder and 
louder, both the adepts and the spectators become excited. A wave of 
heat passes through the courtyard. The women, in their separate room 
or up on the roof, begin to ululate their excitement and their im¬ 
patience to begin the hadra. I have seen women become so excited 
during the recitation of the dikr that they begin to dance as if the 
hadra had already begun. The men disapprove of such behavior, and 
they shove the women roughly out of the court. “The hi zb and the 
dikr are for men,” they often say. In fact there appears to be con¬ 
siderable tension between the men and women at this stage of the 
performance, a tension which, of course, is dramatically symbolized 
by the separation of the sexes. Children, too, become more excited 
and belligerent and are treated equally roughly by the adults and the 
curious teenage boys who inevitably show up at these Hamadsha cere¬ 
monies. The patient, who does not actively participate in the recitation 
of the litanies, is nevertheless moved by the excitement of the group. 
Finally, when the excitement is at a peak—anywhere from 5 to 45 
minutes or more from the beginning of the dikr—the adepts stand, 
hiss out the name of God or the last syllable of His name, “llah,” as 
rapidly as possible, and start to dance the hadra, while the ghiyyata 
blare out their whining tunes and the drummers beat out their 
monotonous rhythms. 


The Hot Part of the Hadra 

The hadra is divided into three principal parts: the hot part (es- 
skhun)i in which the ghita is played along with the tabil and the gwal; 
the cold part ( al-barid ), in which the ghita is replaced by the nira or 
sometimes the ganbri; and the hadra gmaviyya , which uses the instru¬ 
ments of the cold part but is derived from the ceremonies of the 
Gnawa. 

The hot part of the hadra comes first and is the loudest, fastest, and 
most violent part of the performance. The two ghiyyata, a snare 
drummer or two, and three or four gwal players stand at one end of 
the dance area; the other adepts, and any male guest who wants to, 
line up shoulder to shoulder, opposite the musicians and forming an 
outer boundary to the dance area. This line is called the sifa (literally, 



THE HADRA 


193 


configuration) and resembles the dance position of the Berber ahidus. 
The number of men in a line will vary during the performance of the 
hot part, but it seldom exceeds 25. Occasionally, a woman will join 
the line “to cool down.” 

The men pound up and down on their heels, “knocking the breath 
out,” raise and lower their shoulders, and hiss out their breath as they 
seek the hal. They are encouraged by the muqaddim, who dances in 
front of them. He leaps, turns in mid air, lands on his knees, leaps again, 
and stabs at his chest with his fists as he lands. His stabbing motions 
are reminiscent of those of the followers of the Barong who stab 
themselves with krisses in the Balinese dramatic portrayal of the strug¬ 
gle of the forces of good (Barong) and evil (Rangda) (Belo 1960:96- 
124). A few women arc usually drawn by the music of the ghita out 
of their separate quarters and into the dance area, where they bob 
up and down, in trance, or pitch from side to side in front of the ghiy- 
yata. Their feet remain in place; their pitching movement is faster than 
anything possible in a fully conscious state. Often they swing their 
heads so low as to graze the floor with their hair, but I have never seen 
them hit it. They arc generally ignored at this stage of the hadra, but 
if they become too excited they are made to join the line of men. The 
patient, especially if he is a man or boy, is encouraged to participate 
in this part of the hadra. He is sometimes forced to join the line, or he 
is placed in the center of the dance area and made to dance by the 
muqaddim. 

The audience shuffles about when the hadra begins. Spectators ap¬ 
pear to take the positions that are most satisfying to them. Some prefer 
to huddle near the ghita players, others to stand behind the line of 
men, and still others to remain as far back from the dance as possible. 
The basic dance pattern is illustrated in Figure 10. 

From time to time, the men in the line will chant short phrases or 
simply the name of God. The most common of the phrases is “Allah! 
Allah! Allah the Eternal! Allah the Adored!” which Sidi *Ali is said 
to have chanted whenever he performed the hadra. Other popular 
phrases are: “By the grace of Mohammed”; “Allah, our master, it is 
the door, it is the door to my refuge in Allah the eternal”; “O Allah, 
our master, it is the dearest door of communion since the commemora¬ 
tion of the celebration of the birth of the Prophet”; “O Sidi £ Ali, help 
the tardy. O ben Hamdush, be generous with your satisfaction with 
me”; “You are the possessor of safety. Help me, O son of Lalla e ATsha, 
help me”; “O you who have abandoned me, denuded of spirit, whose 
love has embraced my heart”; “I seek your help, O Moulay Abdelqa- 



194 


THE HAMADSHA 



gh ghita player 
gw gwal player 
tb snare drummer 
mq muqaddim 
m man 
w woman 
audience 

Figure 10. The basic pattern of the hot part of the hadra. 

der Jilali, my protection. Come to my aid. I shall carry myself to Tadla 
to make a pious visit to you, O Bu'abid, the nobility of my eyes”; “O 
Prophet of God, nothing can save us from the fires of hell”; “Aloulay 
Idriss, the house of safety”; “O saints, whose role is to bring aid and 
assistance”; “Door of doors. The house here below is the life of error 
and waywardness. Do not build palaces, because you shall move and 
abandon them”; and “There is no God but God, my helper. I pray 
to God the generous. He, my eternal master.” There is no order to 
these phrases, and their meaning often remains obscure. I have been 
told that they are the phrases the ghita speaks. “You hear them just as 
you see things in those pictures you showed me,” one of my informants 
explained, referring to my Rorschach cards. 




THE HADKA 


195 


Trance and Dance Patterns 

The hadra is designed to lead the performers, including the patient, 
into an altered state of consciousness, or trance. 5 The Hamadsha rec¬ 
ognize two such states: hal and jidba. Hal , which means temperature, 
condition, or state, and is in general use in Moroccan Arabic, has, as 
I have noted, a specific meaning in the Sufi lexicon. It refers to one of 
the psycho-gnostic states (, ahaaval ), such as nearness to God or divine 
intimacy, over which the mystic has no control. These states are 
descended from God. In at least the popular brotherhoods of Morocco, 
hal refers to the entranced state which in the Sufi lexicon is known as 
wajd, or ecstasy. Insofar as the Moroccan trancer has no control over 
his hal—it is descended from a saint or a jinn and ultimately from God 
—the hal resembles the ahwal of the Sufis. 6 

For the Hamadsha, hal is both a generic and a specific term. Generi- 
cally, it refers to any trance which occurs during the hadra; specifi¬ 
cally, it refers to a nonviolent trance, which corresponds roughly to 
what is known in the literature on hypnosis as a somnambulistic state. 7 
It is usually attributed to the saint, and may be preliminary to the more 
violent jidba. Its semantic field probably varies slightly with reference 
to men and women, for it seems to encompass a deeper trance and 
more frenetic behavior when used for women. This may result from 
the fact that women fall into trance with much more ease than men, 
and their initial dance steps are more violent. 

Jidba, which is a more frenetic trance, is derived from the Arabic 


8 Trance shall be loosely defined as a complete or partial dissociation, character¬ 
ized by changes in such functions as identity, memory, the sensory modalities, 
and thought. It may involve the loss of voluntary control over movement, and 
may be accompanied by hallucinations and visions which are often forgotten. 

• It is curious, however, that in Morocco a person in trance is said to be “out of 
his conditions” (kbarij al-ah'ival ); a lunatic is also said to be “without his hal” 
{bila hal). To enter hal may be rendered tla* fib l-hal (the hal mounts in him) or 
jab l-bal (the hal arrives in him). The second expression may also mean to 
ejaculate (Merrier 1951). The connection between hal and the mental state at 
the moment of ejaculation is not made, however. Thayyar —which is related to 
words meaning bewildered, confused, or dumbfounded ( bayr ) and anxious, 
angry* and confused, mhayyar —is also used with reference to entering trance. 

T During the somnambulistic state, quite complex behavior can be carried out 
—“behavior which is ‘unconscious’ but nonetheless subject to control, which 
clearly bears the imprint of the culture to which the individual belongs, and which 
is modified according to the makeup of the individual personality, the positive 
impulses which fit in with the trance role he is playing, and the negative or 
inhibiting restraints which the remainder of his normal personality, although in 
abeyance, is able to exert” (Belo 1960:212-213). 



196 


THE HAMADSHA 


for attraction. It is used by the Sufis for the mystical attraction to 
God. Although the transition from hal to jidba is always abrupt and 
dramatic, jidba does not altogether correspond to such “strong sei¬ 
zures” as are described for trance in Bali (Belo 1960:213). It is, 
as we will see, a structured experience in which the performer carries 
out relatively complex behavior, but does not appear to have the 
same control as he does in hal. It is comparable to a state of extreme 
rage (cf. Belo 1960:223). Except in its most extreme manifestations, 
it does not exhibit the universal features which come “from too deep 
a level to be influenced by custom or by idiosyncratic trends in the 
personality” (Belo 1960:212). 8 Jidba is usually attributed to ‘Ai'sha 
Qandisha or another jinn, and not to the saint. It may occur in any of 
the three parts of the hadra. 

Men and women differ significantly in their trance and dance pat¬ 
terns. 9 Women, who fall into trance much more readily than men do, 
remain in the audience until “they are driven by the music of the ghita 
into hal.” They may either work their way through the audience to the 
dance floor and begin to dance in front of the ghiyyata or fall to the 
ground, unconscious and paralyzed, and remain in this cataleptic state 
for a few minutes before they rise and start to dance. 10 The audience, 
whether in the intimate private ceremonies or in the more impersonal 
public ones, is always ready to allow women (or men) to move through 
it to the center dance area. Women who are driven to the center can 
easily be recognized by the ease and even grace with which they manip¬ 
ulate their way through the crowd; they appear to float. Their move¬ 
ments are slow, their eyes are glazed and transfixed, and they seem 
oblivious to all but a “goal” in front of them, in the dance center. 
They are in a light trance. The moment they reach the dance floor, 
they move as close to the ghiyyata as they can—as though seeking 
stimulation—and begin bobbing up and down and swaying from side 
to side in their stylized way. They may continue to wear the veil while 

•Belo (1960:212) argues that while the affective interpretation or coloring of 
trance experience determines the “mood” in which the somnambulistic trance 
activity takes place, the “cataclysmic discharge of nervous tension” of the strong 
seizure is not affected by such interpretation. Its features remain constant, despite 
the emotional background, be it of anger, fear, or sexuality. 

* I have observed on several occasions male transvestite dancers shift back and 
forth from male to female steps, apparently unable to decide on the appropriate 
style. 

10 1 did not see enough performances with the same female performers to note 
whether these two preliminary trance patterns varied with the same trancer or 
with their hearing different musical phrases. 



THE HADRA 


197 


dancing, or they may remove it. This is one of the few occasions on 
which the veil can be removed in public. 

Although women who fall to the ground before dancing give the 
impression that they do so immediately and abruptly, I have found 
that they are often in trance before their collapse. This is not unusual, 
since many members of the audience, especially people closest to the 
ghiyyata, fall in and out of trance throughout the performance. I, too, 
have felt myself momentarily lose contact with what was going on 
around me. It is possible that the women who do collapse either have 
resisted dancing, as may be the case with certain teen-age girls who 
attend school and consider the hadra backward, or are not yet fully 
aware of the prodromata of their trance dancing, as may be the case 
with young and inexperienced trancers. My general impression is 
that the majority of women who collapse before dancing are younger 
and less experienced trancers than those who make their way through 
the audience. The first trance experience of every woman—and every 
man, for that matter—that I have heard about or seen has been sudden 
and abrupt and has involved a collapse. 

Before falling to the ground, the female dancer appears semi-con¬ 
scious—sleepy, one might say—and disoriented. Her eyes are glazed 
and unfocused, and she seems to be oblivious to her surroundings. She 
may nod her head slightly to the rhythm of the music. If someone 
speaks to her, she acts as though she were suddenly awakened from a 
deep reverie and appears to be slightly irritated by the interruption. Her 
collapse is always sudden and involves a loud thud, which attracts a 
good deal of attention. She remains on the ground for several minutes, 
in a cataleptic state which may have been preceded by convulsive 
movements. Sometimes, during the paralyzed phase, she exhibits a 
slight tremor of the hands or feet. Her breath is shallow, her hands 
clammy, and her body cold. If she is left alone, she will gradually start 
responding to the music and will either rest on her knees or stand and 
dance. Often she will begin to dance before her whole body has re¬ 
covered from paralysis. It is not unusual to sec a woman dancing with 
her left forearm and hand still stiff. The woman who collapses is, 
however, rarely left alone. Rather, as soon as she shows any sign of 
life, she is helped to her feet by a woman in the audience or by one 
of the female caretakers from the zawiya, who may dance with her 
for a while until she is able to go on alone. 

The dancer continues to dance in the stylized manner until the 
music ends or she herself is exhausted—“until her jinn has had enough.” 



198 


THE HAMADSHA 


She follows the rhythm of the music and is capable of loosening her 
clothes or removing her veil. Sometimes a woman in the audience will 
tie a kerchief around her chest to keep her breasts from flopping. 
Often her hair will loosen and fall in front of her face or swing 
around it in long snake-like curls. Loosened hair always makes an im¬ 
pression on the men in the audience. Often, too, the hood of her 
jallaba will fall down over her face. No effort is ever made to remove 
it. In this state of hal, the woman will usually exhibit a slight exoph¬ 
thalmos. Her breath will be rapid, her face flushed, and her move¬ 
ments, again, much more rapid than would be possible in a normal 
state of consciousness. 

Sometimes a woman will suddenly fall out of rhythm—this is said 
to happen when her special rih is played—and enter a frenetic trance 
that is called jidba. She may charge around the dance floor, screaming; 
she may act as though she is seeing things in front of her; she may fall 
into a convulsive state; she may suffer temporary paralysis of a limb 
or severe contraction of the face; and she may scratch at her head, or 
even demand a knife and slash at it, with more abandon than the men. 
Sometimes she will use the knife to cut her forearm—usually the left 
one. 11 These acts of self-mutilation are rare among female dancers, 
and especially rare at the ceremonies of the zawiya members. During 
the jidba, the woman’s eyes will also exhibit marked exophthalmos. 
She will appear transfixed. Her breath will be very rapid, almost rasp¬ 
ing, her face flushed, and her movements abrupt and disorganized. 

Certain women in jidba arc compelled to imitate pigs. They begin 
to squeal and grunt and grovel on the ground, and can only be ap¬ 
peased when a little water is sprinkled around them. Such a woman 
is looked on with a certain disgust by both male and female members 
of the audience. She is often said to be marja (lit., bog or swamp), to 
have a wet vagina, possibly from leucorrhea. 12 It has to be remembered 

11 1 have on at least one occasion seen a woman try to cut off her breast. Female 
worshippers of Cybele cut off their breasts just as her male devotees castrated 
themselves (Weigert-Vowinkel 1938, quoting Henri Graillot, Le Cult de Cybele, 
Mere des Dieux, a Route et dans VEmpire romain, Paris: Fontemoing, 1912). 

11 A woman with a moist vagina is regarded with disgust. It is impossible to 
determine whether the moisture referred to is the normal moisture of the vagina 
during intercourse or a pathological condition. Given the fact that most of my 
male informants preferred to ejaculate as rapidly as possible after penetration and 
did not stimulate their women first, it is possible that they were referring to the 
vaginal secretions of an aroused woman. It is said that a woman becomes a marja 
if ner husband ejaculates when he deflowers her (and, sometimes, for three days 
afterward). 



THE HADRA 


199 


that pigs are considered lowly animals and their meat is taboo to the 
Muslim (cf. Belo 1960:223). 

Most women will stop dancing before the music ends. Some will 
continue to the very end of the music and stop then without any diffi¬ 
culty. They may appear disoriented for a few minutes, but that will 
be the only visible aftermath of their trance. Others, however, will 
not have had enough when the music stops, and these women will fall 
to the ground in a state of paralysis or remain in a cataleptic, often 
grotesque, standing posture until the music has begun again. This state 
is called yabis } and is associated with dryness, as of a plant. It is said 
that the jinn of such a woman has not had enough. She exhibits the 
symptoms of someone who has fallen before beginning to trance. 
If the musicians do not begin to play again immediately, she is usually 
covered with a blanket by the other women. She will remain paralyzed, 
sometimes for hours, until the musicians play her rih. It is said that 
if her rih is not played, she will remain paralyzed or at least suffer a 
general malaise that is often described as “pinching bones.” If she is 
never given the opportunity to dance again, it is believed that she will 
die and that the musicians and her family will suffer dire consequences. 

Although some of the young male dancers, particularly at their 
first trancing, enter hal in the same way the women do, most male 
trancers work themselves into the trance slowly. They join the line of 
dancers voluntarily, and then, by means of the breathing and the dance 
techniques, work themselves into the hal. In hal, they are able to fol¬ 
low the rhythm and dance steps of the hadra; their movements may 
be simplified and appear graceful and even dreamy. Their eyes are 
fixed and often exhibit slight exophthalmos. Their limbs may quiver 
slightly or become stiff. Their speech is slurred, and they seem dis¬ 
oriented. They are able to fall out of trance with ease, especially if 
the situation demands their conscious attention—as, for example, at¬ 
tending to guests if one should happen to be host. Subjectively, in¬ 
formants report a loss of temporal and spatial orientation and of body 
consciousness, except in the head, which feels swollen and warm and 
may itch slightly. They hear nothing but the music of the ghita, and 
they see shadows (singular form, khayal) before their eyes. Often they 
remember nothing of the experience. 

My body tightens. It becomes more and more tight. Then I throw myself 
to the ground. I feel hot. I see only c Aisha during the hadra. She is in front 
of me. She too dances the hadra. My head itches. My eyeballs do not move. 




200 


THE HAMADSHA 


I am not conscious of my body. I sweat a lot. I do not know where I am 
or what time it is. I remain this way for about two minutes and then I 
begin jidba . 13 

From time to time, one or two of the male dancers will leave the line 
and begin to dance in the center in a more frenetic fashion. Or he 
will fall to the ground in a cataleptic state, followed occasionally by 
convulsive movements and violent tremors of the limbs. Then it is said 
that the dancer has heard his special rih and has fallen into jidba. 
Gradually, as the musicians continue to play a particular tune to which 
he is responsive, the dancer’s body will relax, and he will stand and 
charge wildly around the dance area, screaming and scaring the 
women and children in the audience. 14 At this point he is unable to 
follow the dance rhythms, although he may become obsessed with the 
ghita and push at it or beckon it seductively. Sometimes he will stimu¬ 
late his genitals with his hands or by rubbing himself against another 
dancer of the same sex. 15 He may scratch his head, and occasionally 
he will ask for a pocket knife and slash at his head with it until his 
face and shoulders are drenched in blood. The knife is held so that the 
blow is softened by the dancer’s thumb. These wounds are usually of 
a superficial nature and do not reach the cranium, although most slash¬ 
ers report having mutilated themselves at least once in the course of 
a lifetime with more abandon than usual. 16 

“ Reported by an experienced Hamdushi in his middle thirties. 

14 Those dancers who charge around most violently, simulating attack, usually 
end up slashing their heads. Aggression is first turned outward but controlled ana 
then turned toward self with less control. This pattern is also found among the 
Balinese kriss dancers. Belo (1960:3) has called attention to parallels with the 
amok psychosis. This pattern of aggression turned inward is not infrequently 
seen in the sudden transition from rage to depression found among at least those 
Moroccans of the Hamadsha’s milieu. A man who, for whatever reason, is enraged 
and exhibits all the overt symptoms of his anger, suddenly retires to a comer of 
his house and sits there silent and immobile for hours if not days. Such a man 
is described generally as mqallesh and, significantly, by the Hamadsha as mqan- 
desh. “Mqandesh,” they say, is derived from “Qandisha.” Several of my head¬ 
slashing informants were prone to such depressions. 

“None of my male informants have ever reported having had an ejaculation 
during trance. The Hamadsha ignore sexual behavior during trance, and they 
make no comparison between trance and the sexual act. Indeed, they are shocked 
by the very suggestion. Some women have remarked, however, that “the feeling 
after a ‘good’ trance is like the feeling you have when you have made love, and 
no one has interrupted you.” 

“It is possible that the wounds then reach to the cranium. I have seen several 
Hamadsha with large cup-like indentations in their skulls which resulted from a 
blow with a club. Hamadsha who have hit themselves on several occasions with 
such abandon often seem punch-drunk. It is said that some Hamadsha have killed 



THE HADRA 


201 


Traditionally—and today during the musem—the Hamadsha slashed 
their heads with a single-bladed iron axe ( shaqur ) or a double-bladed 
one ( shaqria) that resembles the medieval halbard and is not dissimilar to 
the axes found at Cnossos. They also hit their heads with an iron ball 
(kura) often weighing as much as twenty pounds, with a club (zar- 
ivata ) covered with nails and also weighing as much as twenty pounds, 
and with bundles of sticks. The Hamadsha are proud of these instru¬ 
ments of mutilation and handle them with great care. 17 

Some male dancers in jidba imitate camels; they walk stiffly to¬ 
gether in pairs, and then move apart and battle each other as camels 
often do. They make deep, rasping sounds which are likened to the 
grunts of camels. 18 A few men imitate pigs. 

Subjectively, the dancer in jidba reports the same symptoms as in 
hal, but more extreme. He hears nothing but the sound of the ghita 
and the blood throbbing in his head, which feels ready to explode and 
itches intensely. This itching sensation has been described as “like a 
wasp under the skin.” 19 When it occurs, the dancer may see 'Aisha 
Qandisha before him, slashing at her own head with a piece of iron 
and compelling the dancer to do the same. Most of my informants who 
had experienced jidba could recall nothing of the experience. 

I fall to the ground then. If I am made to get up right away then I can 
tell you what I have seen exactly. If I am not forced to get up, but get up 
on my own, then I do not remember what I have seen. I am hot and 
breathe heavily. I feel myself throbbing. There is much itching. I am not 
conscious of my body. I do not know where I am. Nor do I know what 
time it is. My body feels like boiling water. It is frightening. I see only 
‘Aisha. [How do you feel during the head hitting?] It is e Aisha who makes 
me hit my head. I see her in front of me. She has a piece of iron. She is 
hitting her head. I begin to hit my head also. I am not aware of the fact 
that I am hitting my head. There is itching and sweating, and my whole 
body is hot. When c AIsha stops hitting her head, so do I. Then I continue 
to dance. My body and head continue to throb. I am not aware of my 
wounds. I am still hot. I am sweating a lot, and my breath is very fast. 20 

It is said that it is the itching that makes the dancers slash, and the 
flow of blood that calms ‘Aisha. The Hamadsha often smear blood 

themselves by hitting their heads with a jug that cracked; a shard penetrated to 
the brain. 

1T The instruments may be seen in the Musee de l’Homme in Paris. 

“Camels are often castrated to make them more tractable. 

“ c Aisha Qandisha is sometimes said to take the form of a wasp. 

“Cf. fn. 13. 



202 


THE HAMADSHA 


from their head wounds on ailing parts of a patient’s body. Wester- 
marck (1926(1):203) reports that they dipped cubes of sugar into the 
blood and gave it to the sick to eat. I have heard that they also dipped 
bread in their wounds and fed it to the ailing. In those instances in 
which ‘A'isha is not “seen” dancing in front of the head slasher, she is 
considered actually to have entered his body and lodged herself in 
his head. She is responsible for the itching. There is no consciousness of 
pain during jidba, immediately afterward, when the dancer has 
“cooled down” and is again in hal, or after the dance itself. Often 
dancers do not know that they have wounded themselves until they 
discover bloodstains on their clothes. They do nothing to help the 
wounds heal. 

Hot Part of the Hadra Falls 

The hot part of the hadra may last from five or ten minutes to over 
an hour, depending on the mood of the musicians and the condition 
of the dancers. Usually it ends unexpectedly with several long blasts 
of the ghita. The hadra is said to have fallen. Most of the dancers 
will return to their places in the audience. These dancers appear to 
come to their senses with considerable case, although they may be dis¬ 
oriented for a few minutes. Some dancers do not return to a normal 
state of consciousness, but remain in a sort of daze until the cold part 
of the hadra begins. There are usually a few dancers who do not re¬ 
cover from the trance at all and will stay paralyzed on the floor or in 
a standing position. By playing their rihs, the musicians will make 
them recover their senses during the cold part of the hadra. The 
Hamdushi whose experiences I have been quoting describes his sensa¬ 
tions at the end of the hot part of the hadra this way: 

The moment the hadra stops, my body begins to tremble. Sometimes I fall. 
I am still not conscious of having hit my head. There is no itching. I have 
cooled them [the jnun] down with the axe. 1 am not conscious of my body. 
I still do not know where I am or what time it is. I can repeat the fatha, 
but I do not know what I am saying. It is the body orifices that speak—the 
children of ‘Alsha. They are in my body when I make the hadra, and 
c ATsha is in front of me. There is one of them on each of my shoulders. 
I am not aware of the words I am saying. I do not remember them after¬ 
wards. My body is still hot. 

This description is somewhat confused. My informant seems to have 
identified the “children of ‘Alsha” inside him with the two angels who 




THE HADRA 


203 


are said to keep count of the good and bad deeds a man commits each 
day. 

The musicians on one side of the courtyard and the men in line on 
the other form the boundaries of the area in which the action of the 
hadra takes place. They set off the dance area from the audience and 
transform it into a stage. On stage there is a tension, in dramatic 
terms, between the two groups. Women dance individually in a less 
controlled fashion next to the ghiyyata; men, until jidba, dance 
together in a more controlled manner. The ghita and the drums excite 
the dancers and lead them into trance. “The ghita talks to me” and 
“The ghita speaks to my jinn” are common descriptions of the ghita’s 
effect. The line, the most distant point on “stage” from the ghiyyata, 
is said to cool down those who entered jidba. 

The ghita draws men from the line, as well as members of the audi¬ 
ence, into the center of the stage. Here they dance individually, often 
fall into jidba, and sometimes slash their heads. Occasionally the musi¬ 
cians—especially the drummers—are drawn into the center too. The 
muqaddim, whose principal function at this stage of the hadra is to 
encourage trance, may also fall into jidba and lose control of his 
dancing. The patient is encouraged to fall into jidba, and exhibits the 
symptoms described above for other dancers. It is not necessary for 
him to slash his head in order to be cured, although the experience of 
jidba is usually necessary. Sometimes the muqaddim or a sharif, if a 
sharif should happen to be present, will walk on the patient’s back 
or spit on him. £ Aisha Qandisha is said to be under the ground at the 
center of the stage, and it is from there that she will emerge to possess 
an individual dancer or to dance in front of everybody when her par¬ 
ticular rih is played. ( c Ai'sha’s rih is played later, during the cold part 
of the hadra.) It is £ Alsha’s presence that accounts for the fact that the 
Hamadsha must dance barefoot and that they become enraged and 
tear at anyone who wears new slippers and walks in them across the 
stage. 21 Her presence also explains why some dancers fall to the 
ground and kiss the dance floor. 

*1 have never seen or heard of a case where the Hamadsha have attacked a 
person in slippers. Members of the audience are careful not to wear new slippers 
in front of the Hamadsha. In fact, one of the most important jobs of the female 
caretakers is to take the guests’ shoes and put them away. It should be noted that 
the ‘Isawa in trance attack and tear apart any one wearing black. This occurs 
frequendy at the musem. Women frequendy wear black at this time, despite their 
knowledge of its possible consequences. 



204 


THE HAMADSHA 


Music of the Hadra 

The music of the hot part of the hadra is sometimes called the music 
of the father, and it is said to have the color white. The ghita is the 
hot part’s most important instrument. Drums play the constant, mo¬ 
notonous background music, but the ghita—and in other sections of 
the hadra the nira and the ganbri—carries the highly ornamented 
melodic phrases which drive the participants into trance. The rihs 
of the hot part of the hadra are classified into at least four groups, 
each of them named for a tribe: Zerhuni, Hasnawi, Gharbawi, and 
Bukhari. The Zerhuni group is sometimes divided into an £ Allaliyyi 
and a Dghughiyyi subgroup; the Gharbawi has one major subgroup, 
the Khalifi, named after a saintly lineage of the Gharb called the 
Wulad Khalifa. The phrases are also divided according to old and 
new rihs, but this division appears to be purely personal and refers 
to the power of the rih. The older the rih, the more powerful it is 
said to be. 

Individuals are said to respond specifically to certain rihs. One rih 
may leave a man unaffected, a second may drive him into trance, and 
a third may cause him to be possessed by a jinn he follows. I know 
of one man who claims to follow four different jnun—he is a member 
of one of the bidonville teams—and who responds to at least four 
different rihs in four different ways. His dance style and his trance 
behavior vary with the phrase played. One rih makes him dance in 
much the same way the muqaddims do; a second sends him into a sort 
of dreamy trance in which his movements slow down considerably; a 
third drives him into a frenzy during which he dashes around madly, 
eyes protruding, until he spots a glass, chews it to pieces, and swallows 
it; the fourth rih causes him to slash his head. Once a man’s rih is 
played and he falls into a trance, the musicians must continue to play 
variations on that rih until the man has had enough—“until his jinn 
has been satisfied.” If the rih changes before he has had enough, he 
will usually end up in a cataleptic state, as we have noted, until his 
rih is resumed. Then he will dance “until his jinn is satisfied.” 

The phrases of the hot part of the hadra do not, however, have 
specific and generally accepted associations with specific named- 
jnun. They are all vaguely associated with £ Aisha Qandisha, but they 
have no color and incense associations. Some Hamadsha relate one or 
more of them to specific jnun they follow, but these tend to be per¬ 
sonal associations. They are perhaps modeled on the elaborate jinn- 
rih associations of the hadra gnawiyya. In this section of the hadra, 



THE HADRA 


205 


individual rihs within the named groupings do not have names. They 
are indicated by either humming or by singing them with the word ba 
(father). Sometimes they are referred to by the words that are chanted 
when they are played, but this sort of reference tends to be very per¬ 
sonal, since the phrases are not necessarily associated with any particu¬ 
lar chants. 

There does not seem to be a fixed order to the musical phrases. 
The adepts of one zawiya usually begin with one order and those of 
the second with another, but the order soon changes and will vary 
from performance to performance. The choice is made by the ghiyyata 
while they play. There are no pauses during the hot part of the hadra, 
and the ghiyyata move from rih to rih and elaborate on them all, in 
much the same way that a group of jazz musicians will improvise at 
a jam session. The decisions are not fully conscious and seem to de¬ 
pend both on the mood and whims of the ghiyyata, the condition of 
the audience, the needs of individual dancers—including the patient, 
if he should happen to be dancing—and perhaps on certain structural 
necessities within the music. 22 The ghiyyata manage to maintain a 
high emotional pitch throughout the performance. I have found in my 
notes, for example, that whenever I remark that I—or the audience— 
am getting bored, a notation almost invariably follows indicating a 
change of tune and consequently the appearance of a new trancer who 
is driven to perform some dramatic feat such as slashing his head or 
sitting on hot coals. It is impossible to determine to what extent the 
ghiyyata, who are familiar with the trance patterns and musical phrases 
of many of the performers, choose to play the phrase, and to what 
extent the performer himself responds to the boredom and distraction 
of the audience by seeking attention. 

The musicians take the major responsibility for leading dancers into 
trance, but they are not Svengalian puppeteers who control the emo¬ 
tions, the trancing, and the behavior of the dancers with only an eye 
to the dramatic quality of their performances. They are concerned 
about the individual dancers, and will refrain from playing a person’s 
rih if they feel he ought not to dance that night because he has, say, 
been sick with dysentery or the flu. They also respect the wishes of a 
person’s friends and relatives who may ask them not to play his rih 
for one reason or another. Finally, they never stop before a trancer 
has had enough. 

The first section of the hot part is called trish l-hadra } or the slap 


The music has not yet been analyzed. 



206 


THE HAMADSHA 


of the hadra . 23 It has been described to me, not in the musical terms 
I expected, but in terms of its effect upon a participant. 

It is the beginning of the hal. The body of the dancer feels caressed. His 
body is swollen and itches. Slowly he becomes absent from his conditions 
[kharij l-ahawal]. There are “degrees” which climb slowly to the head. 
When they reach the head, the dancer no longer feels his body. It is like 
being drunk. You no longer know your father from your mother or brother. 
Your eyes appear normal. You no longer see anything but darkness. You 
breathe a little faster. The person looks as though he is fainting. He does not 
swear. He does not see the jnun; he can still distinguish people. You can 
tell the notes apart. You are aware of where you are and what time it is. 
The head is dizzy. He is not yet in jidba. There is no head hitting at this 
stage. [I had asked a few questions toward the end of this report .] 24 

The Hamdushi who gave this description was referring to a man who 
danced in line. 

The second section of the hot part of the hadra is called nvast 
l-hadra. My informant described this stage of the hadra as follows: 

There is still the hal, but now he is in jidba. He no longer sees anything 
but the musicians. The ghiyyata are also in jidba. You can tell because they 
begin to sway back and forth. The head throbs. The eyes are open but 
cannot see anything. The breath is fast. You sweat. You are no longer 
conscious of your body even if you hit your head with an axe. Your muscles 
are swollen. There can be head-hitting. You see ‘Ai'sha, who dances the 
hadra in front of you. Her hair is falling all over. 

The ‘Allaliyyin of Meknes call the third section the khammari. I 
have been told that this describes a rhythmic group like the Bukhari 
and the Gharbawi. Non-Hamadsha suggest that the word is derived 
from khamr, or wine. 

The dancers are deeper in jidba. There is even more head-hitting. Six axes 
at a time! The head is stiff and itches. It palpitates. The person’s eyes are 
wide open. You do not know where you are. You only know that c Aisha 
is in front of you, and that if you do not dance she will scratch out at you. 
She has nails like human fingernails. She is just like a person, except her 
feet are like a camel. 

” “Trish” is related to the verb tarsh , to slap or make deaf. “Matrush” is also 
derived from tarsh , and the expression tarshuh ej-jnun, to be slapped by the jnun, 
is used to refer to someone who is slightly crazy, touched. 

“For this description and the following three, see Fn. 13. 




THE HADRA 


207 


The khammari may be followed by ‘ATsha’s hadra, but usually it is 
not played until the cold part of the hadra. 

You care only for £ Aisha. You are not afraid. Those who are in hal are 
never afraid. The hal is the same at this time as during the khammari. 
‘Aisha does not let people sit at this time. 

It is clear from these descriptions of the hot part of the hadra that 
what is important to the dancer is the subjective experience. My in¬ 
formant’s comments on the objective condition of the dancer were 
probably a response to my own questioning. The hadra is seen as a 
single experience that reaches a climax with the appearance of ‘Aisha 
Qandisha. Dr. Theodora Abel, who has examined the Rorschach re¬ 
sponses of some of my Hamadsha informants, was struck by the way 
they responded to the test emotionally as a whole, as a total experience. 
Dr. Abel suggested that this response may have been analogous to the 
total emotional response of the trance experience. To the individual 
Hamdushi, whether a dancer or a musician, the experience can be 
divided only artificially, and these divisions have more to do with the 
subjective experiences of the body than with the musical sections of 
the hadra. 

The hot part of the hadra ends abruptly. The muqaddim or one of 
the dancers who is still in trance begins to recite the fatha. A fatha is 
said for anyone who wishes to obtain a blessing (bar aka ). Women often 
ask for children or the cure of menstrual pains; men, too, ask to be 
cured of various pains and illnesses. Whenever a fatha is said, the per¬ 
son who requested it must give some money. These donations vary 
from a few francs to 10 or more dirhams. They are the brotherhood’s 
major source of income. Although the host is not necessarily obliged 
to give at this time, he usually offers the muqaddim a comparatively 
large donation (5 to 15 dirhams) for the fatha that is said over the pa¬ 
tient. Sometimes the muqaddim or a sharif will massage the patient or 
spit on him while a fatha is recited. Fathas are said until every spectator 
and devotee who wants to receive a blessing is satisfied. 

The Cold Part of the Hadra 

When the recitation of the fathas is over, the cold part of the hadra 
begins. This part is quieter, and it is danced individually rather than in 
line. There are usually a few women who will dance through the cold 
part in front of the musicians, but most of the performers only have 
to dance for a few minutes before they—or rather, their jnun—are sat- 



208 


THE HAMADSHA 


isfied. When an individual hears his rih, he dances in front of the mu¬ 
sicians. It is at this time that some of the more extreme acts of self- 
mutilation occur. These include slashing the head, drinking boiling 
water, sitting on hot coals, eating glass, and similar practices. They are 
intensely dramatic because the performer, usually one of several danc¬ 
ers, is the center of attention. 

The cold part of the hadra seems specifically designed to satisfy in¬ 
dividual needs. If the patient has not danced during the hot part of the 
hadra, or if his jinn has not been satisfied, he will be encouraged to 
dance and to enter jidba during this part of the ceremony. Different 
rifis will be tried until the one that sends him into jidba is found, and 
it will be played until his jinn has been placated. If he should happen 
to be suffering from a paralysis, his limb will loosen up while his rih 
is played and will remain so if he has been satisfied. Usually he experi¬ 
ences some sort of collapse before his jinn has had enough. If the rih is 
stopped too soon, not only will his limb again tighten but he will usu¬ 
ally fall into a cataleptic state. There is seldom a dramatic ending to 
his dance. He will usually walk off to the side and sit down in a slightly 
disoriented condition that lasts for ten or fifteen minutes. Very little 
attention is paid to him. 

Some of the musical phrases that are played on the nira and the 
ganbri, the instruments of this part of the ceremony, are associated 
with named-jnun. All of them appear to affect only certain individuals 
and do not have the wide appeal of the rihs of the hot part. It is said 
that the individual dancer “cools down” during this part of the per¬ 
formance. 

During the cold part of the hadra, I do the rest of the jidba. It is the small 
surroundings. It is the small hal. I begin to remember myself. My head 
returns to the place where it was. My body relaxes. I know where I am 
and when it is. If I have hit my head, I still do not know it. There is still 
a slight throbbing. My body begins to cool down. When 'Aisha stops the 
hadra, so do I. 25 

The dramatic excitement of the cold part of the hadra usually cul¬ 
minates in the performance of £ A‘isha’s rih: “O £ A‘isha. Rise and place 
yourself in the service of the cause of Allah and the Prophet. . . .” 
Often this takes place just after a dancer has abandoned himself to one 

* I have never witnessed or heard anyone in trance talking in tongues in either 
the madina or shantytown ceremonies. Once I heard a man who was said to be 
possessed by an unknown jinn, and whom the Hamadsha claimed they could not 
cure, talk in an imitation French, a language he did not know. He often talked in 
this language when not in trance. 



THE HADRA 


209 


of the more extreme acts of self-mutilation. The lights are suddenly 
extinguished, and the rih is played and chanted for a few minutes. 
‘Aisha is said to rise out of the ground and dance in front of her de¬ 
votees, many of whom claim to see her at this time. ‘Aisha’s rih is fre¬ 
quently accompanied by screams of terror, and when the music stops 
and the lights go on there are often several children on the floor, 
whimpering and in a state of semi-consciousncss. 

The Hadra Gnawiyya 

The cold of the hadra—it may be interrupted from time to time for 
the recitation of fathas—flows directly into the hadra gnawiyya , which 
is “played for the women.” Usually, in the course of the cold part, cer¬ 
tain phrases are played which throw women in particular into trance 
but which do not seem to satisfy them. A special repertory of rihs is 
then played, until the phrase that will satisfy each particular woman is 
found. These are the rihs associated with specific jinniyyas, like Lalla 
Mira and Lalla Mimuna, and consequently with special colors, incenses, 
and perfumes. Rags or scarves of the color pleasing to a particular 
woman’s jinn are tied to her if she has not come to the night wearing 
that color, and the correct incense is burned. Some women—they, too, 
are often called pigs, although they do not actually imitate pigs them¬ 
selves—will begin to dance just as the ceremony seems to be ending 
and the audience is exhausted. They will continue for hours before 
being satisfied. In organization, the hadra gnawiyya resembles not only 
the dance of the Gnawa but also the last part of the ceremonies of a 
number of other popular brotherhoods: the Isawiyya, the Jilaliyya, and 
the Rahaliyya. 

The End of the Ceremony 

When all of the participants are satisfied, the hadra ends, more 
fathas are recited, and dinner is served. The ceremony seldom lasts less 
than four hours, and I have heard of ceremonies that began at 7 in the 
evening and ended at 7 the following morning. Dinner includes cous¬ 
cous and a stew made from the animal sacrificed before the ceremony. 
The men and the women eat separately. Musicians and honored guests 
are given special portions. The dancers eat with voracious appetites. 
The atmosphere is relaxed, if “the hadra has been a good one”—satis¬ 
fying both individually and generally. After everyone has eaten, the 
musicians say a last fatha for the host and for the patients who, if he has 



210 


THE HAM ADS HA 


danced during the hadra, is usually cured of his paralysis or his posses¬ 
sion. His behavior during the hadra will have been no different from 
that of other dancers, and he is the center of attention only when a 
fatha is said over him. The musicians leave and, after a little something 
has been given to the “poor” adepts, the money is tallied and set aside. 
If a particularly large sum has been collected and the host is known to 
be poor, the musicians may give him a bit to help pay for the per¬ 
formance. 

Male dancers who have slashed their heads do nothing for their 
wounds. If the hadra has been good, the dancers will sleep w'cll and 
awake revitalized the following morning. If they have been suffering 
from paresthetic pains, they will experience at least some relief. In most 
cases of jinn attack, the patient will seem to have recovered if he has 
danced. He may still show some signs of stiffness, depression, and gen¬ 
eral malaise, but these will soon wear off. If the hadra has not been good, 
the dancer will spend a restless night, hearing the music of the hadra 
over and over again, and will w r ake up tired and depressed. It is said 
that he will fall sick if he does not perform the hadra soon again. In 
fact, he usually does perform the hadra again as soon as possible. As 
for the patient, he will probably suffer a relapse if his hadra was not a 
good one. Whether their hadras have been good or bad, most men 
report an inability to have sexual intercourse for at least one night after 
dancing. Although they justify this by their desire to remain pure, 
after receiving the saint’s baraka, it is a source of considerable anxiety 
to them. 

The Bidonville Ceremony 

The ceremonies of the bidonvilles are much simpler in organization 
than those of the madina lodges. The sponsor follows the same pro¬ 
cedure as his confrere in the old quarter of the city; however, the 
muqaddim and the team members play a much more active role in in¬ 
viting guests. They are anxious to earn as much as possible from the 
ceremony, and tend to invite all the devotees whom they know r will 
give generously. It is my impression that many more uninvited guests 
attend the bidonville ceremonies. This seems to follow from the more 
open life-style in the bidonville, the more public character of its cere¬ 
monies, and the higher percentage of devotees in the overall shanty¬ 
town population. Financial arrangements between sponsor and team 
are the same as in the madina, and the cost of a ceremony is on the 
average also the same. The sponsor’s deposit is divided among the team 



THE HADRA 


211 


members, as are all of the other proceeds from the ceremony. The 
ghiyyata receive more than any of the other players. 

The average bidonville ceremony takes place at night, although after¬ 
noon ceremonies, which tend to be shorter and less expensive, are more 
frequent there than in the madina. Often the hot part of the hadra is 
played out of doors, either in front of the house or in the nearest square. 
The music consists most frequently of the rihs of the Gharb. There is 
usually trancing and head-slashing. If possible, the patient is made to 
dance and is much more in evidence than he would be in a madina cere¬ 
mony at this stage of the hadra. Trancing patterns and dance steps are 
the same here and throughout the hadra as in the old quarter. Frenzies, 
however, tend to be more frequent and extreme. 

The hot part of the hadra may be continued indoors, but often it is 
stopped “because the ghita are too loud.” The guests enter the house 
as soon as the last, long blasts of the ghita are heard. Fathas are said for 
the host, the patient, and any guest who is willing to pay for one. The 
men and some of the women sit down in the courtyard. (Although 
women in the bidonville ceremonies tend to remain separate from the 
men, this division is never as extreme as in the madina. In fact, bidon¬ 
ville women play a much more active role throughout the entire cere¬ 
mony.) The guests wait for tea, and they may also be given dates and 
milk, a fried saltless bread said to please the jnun, and black olives, 
which are reputed to drive the jnun away. The break between the hot 
part of the hadra and the subsequent pans is much greater here than 
in the old quaner. 

Once all of the guests are served, the musicians who are sitting to¬ 
gether in a corner of the courtyard begin to play. Sometimes they will 
chant a few words of a dikr to warm themselves up. Their dikr, ac¬ 
companied by the ganbri, seldom lasts for more than four or five min¬ 
utes. The rest of the evening closely resembles a madina ceremony, but 
here the cold pan of the hadra becomes a major pan of the ceremony 
and is much more frequently interrupted for the recitation of the 
fathas. The muqaddim and other team members do their very best to 
extract as much money as they can from the audience. They often earn 
as much as 60 or 70 dirhams in a night. Their fathas are more dramatic 
and often involve massaging, spitting, treading on, or knocking the head 
of the person over whom the prayer is said. Occasionally, a fatha-sayer, 
who is in trance, will predict the future. The hadra gnawiyya follows 
the cold part without any interruption. The rihs played are chosen on 
the basis of the needs of the women dancers. When all of the dancers 
have been satisfied, the meal with which the ceremony ends is served. 



11 


The 

Explanation of 
Therapy 


Therapy consists of a structured set of procedures for moving an in¬ 
dividual from one state, most generally the state of illness, to another 
state, the state of health or proximate health. Concerned as it is with 
the individual as a total being, therapy necessarily involves changes in 
all significant levels of human existence—the physiological, the psycho¬ 
logical, and the socio-cultural. Different therapies appear concerned 
primarily with one or another of these levels, often at the expense of 
the others. Although the nature of the illness itself frequently forces 
the emphasis onto one such level, cultural factors are often equally im¬ 
portant determinants. 

Hamadsha therapy, too, consists of procedures that move an individ¬ 
ual from a state of illness to a state of health. “Illness,” “health,” and the 
therapeutic procedures themselves are grounded in the social and cul¬ 
tural life of at least a certain segment of Moroccan Arab society, and 
probably reflect tensions inherent in that life. Although I have not 
thought it appropriate to the level of analysis of this study to consider 
the origin and nature of such tensions, I have found it necessary to call 
attention to some of the more salient tensions as they are reflected in 
the Hamadsha complex itself. I have referred, for example, to tensions 
between city and country, when considering the relationship between 
the Hamadsha of the old quarter of Meknes and those of the shanty¬ 
towns; and to stresses and strains in the male role, concomitant with 
the segregation of the sexes and the extreme patrilineality and patri- 
potestality of the Arab world, when analyzing the hagiographic leg¬ 
ends. 



THE EXPLANATION OF THERAPY 


213 


The following pages will recall the Moroccan’s belief in the inferior¬ 
ity, treacherous nature, and insatiable sexual desire of women. This be¬ 
lief probably reflects an unarticulated if not altogether unconscious fear 
of women—a fear which finds expression in the segregation of the sexes 
in Morocco and elsewhere in the Arab world, and is probably com¬ 
pensated for by the Arabs’ extreme emphasis on male virility. “Male” 
and “female” are, however, more than labels of sexual identity; they 
refer to a whole complex of behavioral traits and symbolize feelings 
that are experienced at some level of consciousness by both men and 
women and arc not, it is suggested, insignificant in the aetiology and 
conceptualization of illnesses treated by the Hamadsha. 

Hamadsha therapeutic procedures are provided with both a “mythic 
charter”—the hagiographic legends that integrate the therapeutic with 
the “religious”—and a theoretical explanation. Both the “mythic char¬ 
ter” and the theory of therapy can be conceived of as legitimating ap¬ 
paratus which, as such, give authority to the therapeutic system and are 
thereby themselves therapeutically significant. It has been suggested 
that the hagiographic legends reflect tensions inherent in the individual 
Hamdushi and in his social and cultural situation, tensions which may 
in fact be responsible for his illness or for the form it takes. That the 
elements of the theory—saints, jnun, and baraka—are more than logical 
constructs for the explanation of Hamadsha therapy has been implicit 
in my analysis of it. They are images, in Godfrey Lienhardt’s (1967: 
148 et seq.) sense of the word, grounded in the social and cultural 
reality of the Hamadsha, of what the Westerner would call psychic 
reality. 1 They are elements in which I would call the participational 
mode of explanation of illness and therapy. They may be considered 
signs of psychic states and sy?nbols of socio-cultural processes. The 
dual ground of these “images” serves to strengthen the legitimations of 
therapy. e Aisha Qandisha and the Hamadsha saints are more than char- 

1 Lienhardt (1967:147 et seq.) speaks of some Dinka spirits, or Powers, as “images 
evoked by certain configurations of experience contingent upon the Dinka’s re¬ 
action to their particular physical and social environment.” These “images,” he 
suggests, are “the active counterpart of the passive element in human experience.” 
It is perhaps significant that in ordinary English usage we have no word to 
indicate an opposite of ‘actions’ in relation to the human self. If the word 
‘passions,’ passiones , were still normally current as the opposite of ‘actions,’ 
it would be possible to say that the Dinka Powers were the images of human 
passiones seen as the active source of these passiones. 

Reference to a Power then, Lienhardt remarks, is not merely a description of the 
experience but an interpretation. In other words, the experience, say, of guilt is 
immediately articulated, if not instantiated, in an interpretive idiom external to 
the individual. 



214 


THE HAMADSHA 


acters in a story. They are rooted in both individual and social life. 
Moreover, the individual’s conflicts are immediately and necessarily ex¬ 
pressed in terms congruent with social reality. They are recast in uni¬ 
versal terms. This in itself is of considerable therapeutic import. Ther¬ 
apy involves the articulation and manipulation of these images. Theory 
and practice are inseparable within the Hamadsha context, as they 
probably are within the contexts of other pre-scientific therapies. 

Hamadsha therapy is pragmatic and subject to considerable varia¬ 
tion. It treats patients in various conditions of life and health, and it 
should be stressed that there was considerable variation among patients 
whom the Hamadsha successfully cured. 2 In some cases, a crisis situa¬ 
tion triggered a reaction that did not impress me as being deeply rooted 
in the patient’s psyche; rather, the patient seemed to be following one 
of several cultural models appropriate to his situation. 3 Other patients, 
however, seemed to be suffering from hysterical, depressive, or even 
schizophrenic reactions. It is possible that a number of chronic cases 
were kept ambulatory by their intense involvement with the Hamadsha 
and their frequent performances of the hadra. 

What follows is an examination of the elements of Hamadsha therapy 
which are recognized in Western theories of psychotherapy as thera¬ 
peutically effective. The suggestions which I make about these thera¬ 
peutically significant factors must be considered hypotheses based on 
a detailed consideration of the Hamadsha complex; they point ahead to 
a sequel to this study in which they will be tested against family and 
individual case histories. I have given particular emphasis to the social 
and psychological dimensions of Hamadsha therapy, not simply because 
I believe them to be the most significant therapeutically but because of 
the limitations of my own training and research opportunities. Primary 
attention has been placed on ceremonies involving the hadra; the pil¬ 
grimage is considered as a complement to them. 


a Ortigues (1966) also stresses the varying personality types who become pos¬ 
sessed by spirits in Senegal. 

“The response to such crises may be considered an acute anxiety reaction of 
short duration, whose symptomatology may include incoherence of thought and 
delusional and hallucinatory activity. It has been claimed that such reacnons are 
more common in primitive than in advanced societies. They may result from the 
failure to perform customary rites (Wittkower 1971). It must be remembered that 
alleged failure to perform ritual obligations may itself—as I believe is often the 
case for the Hamadsha—be symbolic of some breach in the individual’s personal 
morality (cf. Crapanzano 1972). 




THE EXPLANATION OF THERAPY 


215 


Group Support 

Group support, one of the most frequently cited therapeutic factors 
in “primitive” as well as “modern” psychotherapy, plays an important 
role in the Hamadsha cures. The ailing individual is not isolated and 
ignored, but becomes a patient who is the concern of the whole group. 
The group, which consists of family, friends, and neighbors as well as 
the Hamadsha, not only offers the patient sympathy, encouragement, 
and the hope of cure, but also mobilizes itself to cure the patient of his 
troubles. This mobilization serves to alter the relations between the 
patient and the members of the group themselves. Men and women, for 
example, are in closer, more intimate contact during the preparation 
for a ceremony. It is possible that such a rearticulation of social rela¬ 
tions may help to reduce socially generated tensions that are in part 
responsible for the patient’s condition. Mobilization of the group, more¬ 
over, involves sacrifices of time and money that serve the patient as an 
assurance of the group’s care and concern. A responsibility not to let 
the group down is placed upon him. 

The group also provides the patient with an interpretation of his ill¬ 
ness, or reinforces his own or the diviner’s interpretation, and this acts 
to reduce the anxiety that accompanies ignorance of the illness’ cause 
and cure. Fatna, for example, was convinced by friends and neighbors 
that her child Fatima was sick because she had failed to carry out her 
promise to sacrifice a chicken to Sidi £ AIi. Members of the group often 
advocate such interpretations out of a vested interest in them. They, 
too, have often suffered, or are liable to suffer, from illnesses of similar 
origin. In supporting a supernatural explanation of the illness, the group 
affirms that the patient is not necessarily responsible for his own con¬ 
dition. Feelings of shame, remorse, and “guilt” are deflected. The pa¬ 
tient no longer broods over himself. This technique may be of partic¬ 
ular significance in depressive reactions. Moreover, since the patient is 
not held responsible for his condition, he is able to act on socially un¬ 
acceptable impulses both before and during curing ceremonies without 
incurring public censure. These include the refusal to carry out tasks 
normally expected of him, excesses of self-mutilation, homosexual ad¬ 
vances to other dancers, or acts of autoeroticism. 

During ceremonies the group—the audience—becomes more than a 
backdrop for the patient’s cure. The group actively participates in the 
cure. Its members not only tend to the patient’s needs, but also often 
dance and trance themselves. They serve as models for the patient. 
They too must placate the jnun they follow. Ailing individuals who 



216 


THE HAMADSHA 


are unable to sponsor their own ceremonies attend in the hope of a 
cure, and it is not unusual for the patient to tend to their needs. The 
recitation of the fatha intensifies group sentiment. The patient is made 
to understand in dramatic terms that he is not alone in his condition, 
and that many others who have suffered in the past as he is now suffer¬ 
ing have recovered. This “sharing of illness” also occurs at the pilgrim¬ 
age centers, where the patient encounters many other ailing men and 
women and hears stories of extraordinary cures. 

The ceremonies not only permit Hamadsha within the group to 
dance and trance and work through their own personal conflicts, but 
also serve a cathartic function for other members of the audience. The 
audience, like the audiences of exorcistic seances in Burma, does not 
consist “as might be the case in certain theatrical contexts, of jaded 
spectators, seeking entertainment to shake them from their boredom or 
to permit them to escape their routine concerns” (Spiro 1967:196). The 
audience lives through the conflicts that are symbolically played out 
before them and that undoubtedly have deep roots within their cultural 
tradition. This catharsis, too, may serve to reduce social tensions that 
are in part responsible for the patient’s condition. Other tensions, such 
as tensions between men and women, are expressed in less obvious ways. 
Women in the madina ceremonies, for example, must wait to participate 
until the men have completed their chanting; at the end of the cere¬ 
mony, however, a single woman whose jinn has not had enough can 
keep the musicians playing and the men waiting for hours. 

Patient Participation 

The patient plays a role in his own cure. Often he must help with 
the preparations for a ceremony or a pilgrimage. He must dance and 
trance during the night and, if possible, repeat the fatha for others. 
Should he happen to be host, he must tend to the needs of his guests. 
On pilgrimages, he must carry out various ritual activities such as cir¬ 
cumambulating the tombs of the Hamadsha saints, bathing in the c Ayn 
Kabir, or smearing himself with henna from ‘A'isha’s grotto. After the 
pilgrimage or night, he is often required to follow certain rules. These 
may be stereotyped and appear arbitrary; still, apart from whatever 
symbolic role they play, they give the patient the feeling that he is 
doing something for himself. He is keeping his jinn or jinniyya at bay. 

The importance of patient participation in the Hamadsha cures 
should not, however, be exaggerated. The patient’s role is not as active 
as in psychoanalysis or psychodrama. He is passive before the exorcist- 



THE EXPLANATION OF THERAPY 


217 


seers, who uncover the source of his troubles and suggest a cure; he is 
passive before the saints, the jnun, and even the music of the ghita. 
This passivity can be considered as indicative of the patient’s basic exis¬ 
tential stance toward his world or as a period of socially sanctioned 
regression to a stage of dependency which may be of therapeutic value. 
During the cures, and especially during the pilgrimage, the patient is 
removed from his normal position in society with its concomitant de¬ 
mands and pressures; he is given a respite. 

The patient is not necessarily the center of attention during cere¬ 
monies, although he is the manifest cause for the ceremony. Indeed, he 
is often almost completely ignored. I remember, on several occasions, 
leaving a night only to discover that I had not made a single notation 
about the patient. This lack of attention during the formal curing may 
be of significance in the cures of certain hysterical individuals whose 
dramatic symptoms and life histories suggest that feelings of neglect 
and the desire to be the center of attention are important etiological 
factors for their condition. The group sometimes appears to treat them 
with the same belle indifference with which they treat themselves. 

Patient-Curer Relationship 

Although the importance of the patient-curer relationship has been 
stressed in explanations of the effectiveness of psychoanalysis, hypnosis, 
shamanistic cures, and other therapeutic techniques, I have not found 
this relationship to be particularly significant in the Hamadsha cures. 
I do not mean to imply that the role of the curers—the exorcist-seer, 
the saint’s descendant, or the muqaddim of a team—is unimportant. 
It is important, especially when the curer has a strong, charismatic per¬ 
sonality. I am suggesting, rather, that neither a deep transference rela¬ 
tionship with archaic features nor an important personal bond is estab¬ 
lished between patient and curer. Whatever transference is established 
is immediately deflected to supernatural agents—the saints and the jnun. 
Even the power of the curer is conceived in supernatural terms, in 
terms of baraka. This, in my opinion, is not simply an instance of de¬ 
fensive projection. The saints and jnun are a given in the patient’s 
world, providing him with a symbolic schema for the interpretation of 
his experiences. It is on this level that relationships are manipulated. 
The patient’s relationship with c AIsha Qandisha, for example, is funda¬ 
mentally changed. 

The passive role of the patient is reflected in his relationship with 
the curer. Even when the curer plays an important, personal role in the 



218 


THE HAMADSHA 


treatment of the patient, he does not attempt to elicit confessions from 
the patient or to review his life. He tries to suggest to or persuade the 
patient that he has in fact been attacked by a jinn, that the jinn requires 
a sacrifice or a night, and that the patient, if he follows the commands 
of the jinn, will be cured. The curer’s suggestions are not patient- 
specific. He reiterates what the group and the patient know already or 
at least suspect. 

Insight and Suggestion 

No attempt is made either by the curer or by members of the pa¬ 
tient’s group to give him insight into the nature of his problems. There 
is no psychic probing in the Hamadsha cures. Since the patient is not 
responsible for his condition, any self-examination would be inappro¬ 
priate. Instead, the patient is provided with an interpretation and dy¬ 
namic explanation of his condition. He is urged by the group and by 
individual curers to accept this explanation. Techniques designed to 
increase his suggestibility play an important role in Hamadsha cures. 
Often dreams, especially at the saints’ tombs, will be interpreted in 
terms of the explanations proffered and considered authoritative. 

Cult Membership 

The patient who has been treated by the Hamadsha is not the “same” 
person that he was before his illness. Most Hamadsha cures do not end 
with the pilgrimage or ceremony. In the case of the pilgrimage, the 
patient forms at the very least a strong dependency on the saint and 
perhaps on ‘A'isha Qandisha. Although he does not necessarily become 
a devotee of the Hamadsha, his subsequent experiences will usually be 
interpreted in terms of the saint and ‘Ai'sha. In the case of the cere¬ 
monies, the patient often recovers but is required to follow certain rules 
to please his jinn. Usually he will have to perform the hadra from time 
to time and sponsor an annual ceremony to commemorate his cure. If 
he fails to perform or to sponsor a ceremony, there is a strong likeli¬ 
hood that he will again fall ill. In other words, the patient becomes a 
member of the Hamadsha cult. 

Cult membership provides the patient with a new social status. He is 
now a Hamdushi or a devotee of the Hamadsha. His social relations 
are now articulated with respect to this new status. Others must take 
cognizance of it. A husband whose wife becomes a Hamdushiyya must, 
for example, sponsor an annual ceremony for her and permit her to at- 



THE EXPLANATION OF THERAPY 


219 


tend the ceremonies of other devotees. Membership in the cult enlarges 
the patients’ social network and, at least in the case of the bidonville 
Hamadsha, gives them the opportunity to develop multiplex relations 
in a new and alien environment. Should the patient happen to become 
a member of a team or zawiya, he is entitled to the benefits of this asso¬ 
ciation. 

With a change in social status, there is a change in social identity 
and in self-image. The patient is treated as a Hamdushi or a devotee 
of the Hamadsha. Within the cult, especially if his behavior during the 
hadra is dramatic or if he should happen to know many litanies, the 
patient is considered to have a great deal of prestige—prestige which 
he might not have had before. Outside the cult, he is treated with a 
mixture of awe and respect by other Moroccans of his background. 
He has, after all, been in contact with the jnun and, as a trancer, he is 
not without his share of baraka. However, he is looked down upon by 
other, better educated Moroccans as a poor wretch and a victim of bar¬ 
barous superstition . 4 The patient is well aware of this characterization. 
Within the group he has prestige and outside it humiliation. This hu¬ 
miliation should not be ignored. It may well reflect—and support—the 
patient’s self-image, and be of some secondary gain. 

Membership in the cult also reinforces the patient’s faith in the 
Hamadsha and in their world-view. Again and again, at private and 
public ceremonies, he is given the opportunity to witness their success. 
Attendance at the annual musem renews his faith in the Hamadsha 
saints and their descendants. Finally, and most importantly, he is given 
the opportunity to perform the hadra and as we shall see, to work out 
—in a structured, albeit symbolic, fashion—conflicts which are respon¬ 
sible for his condition. 

Trance, Abreaction, and Catharsis 5 

The Hamadsha consider the trance to be of paramount importance 
for their cures. They say that their patient must dance until his jinn is 
“satisfied.” Usually, to satisfy his jinn, the patient must enter jidba, a 
deep frenetic trance that in many cultures is interpreted as a state of 
possession. The Hamadsha themselves do not necessarily explain jidba 
as possession; it is simply that the patient must enter jidba before his 
jinn is said to have had enough. 

‘The Hamadsha are referred to by many of them as “black”—a possible indi¬ 
cation of lower status. 

‘See the Appendix for factors that appear to induce trance for the Hamadsha. 



220 


THE HAMADSHA 


I would like to suggest that trance is in fact an important element in 
Hamadsha therapy, and affects the individual both physiologically and 
psychologically. The physiological effects of trance and possession, like 
those of shock therapy, have not been worked out. William Sargant 
(1964), who has considered the problem from the perspective of Pav- 
lovian neurophysiology, compares trancing and possession to narco¬ 
therapy, psychoanalysis, shock treatment, leucotomy, techniques of 
religious conversion, and brainwashing. He suggests that all of these 
techniques produce a state of emotional exhaustion and collapse in the 
subject. This collapse involves the erasure of former conditioned pat¬ 
terns and beliefs and a heightened state of suggestibility. 6 

The role of collapse in Hamadsha therapy presents a problem. Some 
trancers fall into a state of collapse before they begin to dance and 
enter jidba; others collapse in the middle of the dance; and still others 
do not collapse at all. No ceremony is terminated with a dancer in a 
state of collapse. It is not altogether clear, however, what Sargant means 
by “collapse.” He suggests that the terminal sleep of the possessed is 
the period of complete inhibitory collapse, but this does not appear to 
be a satisfactory solution. Most trancers recover from a possession state 
well before retiring. They usually show signs of renewed vitality, well¬ 
being, and recovery from the symptoms of an attack by a jinn. 7 

Suggestion does appear to play an important role in any case. This 
applies to dancers and members of the audience as well. 

•Sargant’s work has been criticized for assuming a neurophysiological basis for 
the kind of “collapse state” produced, say, by the fiery preaching of John Wcstlcy 
(Prince 1968:132). Nevertheless, his observations concerning the erasure of former 
conditioned patterns and beliefs, including a tendency in the so-called “ultra- 
paradoxical phase” for such patterns to turn from positive to negative and from 
negative to positive, and the heightened suggestibility of the subject during periods 
of emotional exhaustion, arc of considerable significance. 

7 Prince (1968:131) suggests that Sargant confuses the initial collapse with the 
final sleep of exhaustion. Sargant himself recognizes that the threshold for col¬ 
lapse probably varies with the temperament of the trancer. One might well add 
that it also varies with his life circumstances, including such factors as stress, 
and his experience of the trance. 

Some persons can produce a state of trance and dissociation in themselves, 
or in others, with a decreasing need for strong and repeated emotional 
stresses, until it may become so much a conditioned pattern of brain activ¬ 
ity that it occurs with only minor stresses and difficulties; for example, in 
the primitive religious context, at the renewed beat of a drum, or the scream¬ 
ing roar of the rhombos (Sargant 1964:103). 

Experienced trancers among the Hamadsha enter trance with greater case than 
inexperienced ones. The first notes of their particular rib will often send them into 
a deep trance or into a state of collapse. 




THE EXPLANATION OF THERAPY 


221 


States of possession or trance have also been used by numerous religions to 
try to help the spectator, as well as the possessed person, to accept the 
relevant doctrine as true. If the trance is accompanied by a state of mental 
dissociation, the person experiencing it can be profoundly influenced in his 
subsequent thinking and behavior. Even if the spectators remain unmoved 
and devoid of any emotional excitement, it may still help to persuade them 
of the truth of the belief professed, especially if they have been led to think 
that a trance means that the person concerned is possessed by, or in com¬ 
munication with, a certain god. (Sargant 1964:103) 

As we have seen, members of the Hamadsha audience are often on the 
brink of dissociation, oscillating between normal and dissociated states, 
or in a state of dissociation. 

The period following collapse, whether initial, medial, or terminal, 
is of extreme importance. Not only is the trancer in a state of extreme 
suggestibility but, if Sargant is correct, he undergoes an erasure, or par¬ 
tial erasure, of previously learned behavioral patterns and beliefs. New 
behavior patterns and beliefs may be instilled, or old ones, including 
“neurotic” ones, may be renewed and reinforced. Of paramount im¬ 
portance is the suggestion that the experience of such a state is of thera¬ 
peutic value. The jinn has been satisfied! 

The extreme suggestibility of the individual in either case calls at¬ 
tention to the “learning” aspect of therapeutic procedures like the 
Hamadsha’s. Belief in the Hamadsha and in their lore is instilled, recon¬ 
firmed, and reinforced. The lore includes the Hamadsha’s “psychology” 
and “pathology.” Both of these have a profound effect upon the in¬ 
dividual. 

Psychologies pertain to a dimension of reality that is of the greatest and 
most continuous subjective relevance for all individuals. Therefore the 
dialectic between the theory and reality affects the individual in a palpably 
direct and intensive manner. (Berger and Luckmann 1967:176) 

In other words the socially accepted interpretation of psychic reality— 
the psychology—“realizes itself forcefully in the phenomena it purports 
to interpret.” This point is of extreme importance if, as I suggest, Ha¬ 
madsha therapy involves the manipulation of “images” of psychic re¬ 
ality. 

The erasure of behavioral patterns and beliefs that Sargant postulates 
may be related to the “regressive behavior” often observed in hypnotic 
and ceremonial trance. It has been found that a person who is deprived 



222 


THE HAMADSHA 


for a sufficiently long period of access to “the usual clues for maintain¬ 
ing a normal sense of reality” becomes oversensitive to both internal 
and external stimuli and appears vulnerable to both (Gill and Brenman 
1966:5). This state of heightened sensitivity and vulnerability, corre¬ 
sponding roughly to what the Hamadsha call hal, is often followed by 
“outbursts from the internal world in one form or another” (Gill and 
Brenman 1966:6). This second stage may itself correspond to what 
the Hamadsha call jidba and interpret as possession by a jinn, or at 
least as the result of jinn activity. Such states are considered to be 
therapeutically significant. 

Explanations of the curative value of such “outbursts” have em¬ 
phasized either the positive value of these regressive states—“regression 
in the service of the ego”—or their “cathartic function.” The patient is 
given a socially sanctioned opportunity for the “stormy discharge” of 
emotions rooted in traumatic experiences of the past or in psychic con¬ 
flicts of the present, and he experiences a reduction of anxiety and a 
consequent alleviation or elimination of anxiety-produced symptoms. 
He “works through” or “acts out” his conflicts, or “relives” his trau¬ 
matic experiences. Although the value of emotional discharge has been 
recognized at least since the publication of Breuer and Freud’s Studies 
in Hysteria in 1895, it should be stressed that the mere outburst of 
emotion is not necessarily of therapeutic value (Gill and Brenman 
1966:356). It may indeed have the contrary effect. I would suggest that 
all such outbursts must be structured if they are to be of positive thera¬ 
peutic value, and that the structures must reflect the symptom-produc¬ 
ing conflicts. 

This point needs elaboration—and here I am indebted to L6vi-Strauss’ 
article “The Efficacy of Symbols” (1963a). The cure involves a two¬ 
fold process. First, a conflict originally existing on an emotional level 
or in the unconscious is made explicit. Second, an opportunity for the 
working out or resolution of conflicts thereby made explicit is offered. 
The making explicit of unconscious processses may involve—or claim 
to involve—the uncovering of actual, lived-through experiences, as in 
the case of psychoanalysis, or it may provide the patient “with a 
language , by means of which unexpressed and otherwise inexpressible 
psychic states can be immediately expressed,” as in certain shamanistic 
cures 8 (Levi-Strauss 1963a: 198). In the case of psychoanalysis and sim¬ 
ilar therapies, the rendering of unconscious processes factually articu¬ 
late provides a means for their “free development” and their resolution. 

• This may be related to the degree of interiorization typically encouraged by a 
given culture. 




THE EXPLANATION OF THERAPY 


223 


This development and resolution may be subject to strict structural 
laws, but it always remains on a personal, particularistic level. The psy¬ 
choanalytic patient “constructs an individual myth,” as Levi-Strauss 
puts it, “with elements drawn from his past” (1963:199). In the case 
of shamanistic cures—and Hamadsha cures as well—the symbolic ex¬ 
pression of unconscious processes is followed by a highly structured— 
ritualized—resolution of symbolically expressed conflicts. Resolution 
here is impersonal and universalistic. The patient “receives from the 
outside a social myth which does not correspond to his former per¬ 
sonal state” (Levi-Strauss 1963a: 199). 

Whether or not the structural laws of both types of cures are the 
same, as Levi-Strauss suggests, there is a manifest difference in the ex¬ 
istential stance of the patient vis-a-vis the therapeutic system. In the 
case of psychoanalysis, the patient is active and the therapist is passive. 
In the case of the shamanistic cure, the patient is passive and the shaman 
is active. 9 The shaman is an authoritarian figure, often identified with 
a mythic figure, who “molds” the patient’s unarticulated conflicts into 
a socially given schema. If the elements of the schema are images of 
psychic states, grounded in social and cultural reality (as I believe is 
true for the Hamadsha), then the shaman’s symbolic manipulations are 
of considerable therapeutic importance. In any event, the special nature 
of the dialectical relationship between “psychology” and “psyche” 
probably permits the curer to modify, at least temporarily, the patient’s 
psychic reality. Since suggestion plays a manifestly important role, tech¬ 
niques such as trancing, which increase the patient’s suggestibility, are 
especially important in shamanistic cures. 

The Hamadsha “pathology” and “theory of therapy” not only ex¬ 
plain these “outbursts,” but structure their very expression. Theory and 
practice cannot be separated here. In the simplest terms, 'Aisha Qan- 
disha or whatever jinn is held responsible for the patient’s condition is 
said to take possession of the patient or, in the case of certain men, to 
force him to slash his head. When the jinn has been satisfied, it not 
only leaves the patient but is transformed into the patient’s support. A 
symbolic transformation has taken place which may involve at least a 
temporary restructuring of “psychic reality”—and perhaps, as Levi- 
Strauss (1963:201) would suggest, even a physiological transforma¬ 
tion. 10 

• In Hamadsha cures, where the role of the curer is not as important as in certain 
shamanistic cures, the patient is still a passive actor and not an innovator. 

10 It is assumed in the following analysis that the patient is a man. The manner 
in which klamadsha therapy functions for the female is not considered, because of 
the spottiness of my data concerning women. This has resulted from the strict 



224 


THE HAMADSHA 


The illnesses for which £ ATsha Qandisha is responsible and which the 
Hamadsha are able to cure include paralysis of the limbs, “pinching 
bones,” sudden deafness, blindness, and mutism unaccompanied by or¬ 
ganic lesions, children’s diseases, menstrual difficulties, and barrenness. 
( £ Aisha Qandisha is also able to limit a man’s virility or render him 
impotent, but Hamadsha cures arc not usually effective in these cases.) 
All of the illness caused by £ A‘isha Qandisha prevent the patient from 
actively performing his social roles and concomitant tasks. He is un¬ 
able to meet the demands of a “male-dominated” society like Morocco, 
which is organized along the lines of extreme pa trilineality and patri- 
potestality. The symptomatology of many of the illnesses—paralysis of 
the limbs, pinching bones, impotence—suggests an inability to play the 
male role; barrenness and children’s diseases prevent the continuation of 
the male line. Illness, especially for men, is conceived at some level of 
consciousness as an inability to live up to the ideal standards of male 
conduct. That is, illness is associated with being rendered a woman. 
Feelings of inadequacy, weakness, and impotence are symbolized by the 
female. 

This conception of illness as feminization may indeed be rooted, as 
the analysis of the legendary cycle of Sidi £ Ali and Sidi Ahmed sug¬ 
gests, in the relationship between father and son. In order to obtain his 
father’s baraka—to live up to the male ideal—the son must play the 
passive role before the father (and identify with the mother). If he 
refuses this solution, he finds himself incapable of meeting the ideal. It 
seems likely that many but by no means all of the illnesses for which 
the Hamadsha are called in have such an origin. Certainly—and here I 
am stepping out of the limits set by the level of analysis of this study— 
the often arbitrary, harsh, and inconsistent behavior of the Moroccan 
father toward his son would seem to preclude the possibility of the 
son’s strong and positive identification with him. It provides the son 
with the ideal but demands behavior contrary to the ideal. The son 
must remain passive—unmanly—before the father. He must be as 
woman to the father. 

Women are stereotyped not simply as weak and helpless, but as 
treacherous, whimsical, untrustworthy, unfaithful, and sexually insa¬ 
tiable. 11 They are the image of 'A'fsha Qandisha and must be pacified 

segregation of the sexes in Morocco. It should be pointed out, however, that 
women rarely slash their heads. The jinniyyas appear to function as supporting 
doubles for them. They arc capable, however, or turning on their female fol¬ 
lowers, and may serve as an externalized conscience. 

“This aspect of the figure of the woman suggests to the psychoanalytically or¬ 
thodox that the mother has in some manner betrayed the son to the father. 



THE EXPLANATION OF THERAPY 


225 


and controlled. c ATsha Qandisha and the other female jmm are, at least 
for the majority of Hamadsha, the cause of their illnesses; they must be 
placated and transformed. This—the cure—is accomplished by means 
of the saints’ baraka. 

The cure involves an attempted resolution of two opposing figures 
—'Aisha Qandisha and the Hamadsha saints. As they are described in 
legend and folklore, they represent a number of different elements 
within the Moroccan’s experiential world and are commonly, though 
less consciously, associated with still other elements. These are sum¬ 
marized below; the less conscious associations, which have been derived 
from life-history material not examined in this work, appear in paren¬ 
theses. 


C AISHA QANDISHA 


SIDI e ALI 

AND SIDI AHMED 

Jinn 


Saint 

Earth 

Water 

Sun 

(low) 

Green 

(high) 

Dark 


Light 

Black-red 


White 

(left) 


(right) 

(interior) 


(exterior) 

Female 


Male 

Blood (menstruation) 


(semen) 

Infertility 


Fertility 

Impotence 


Virility 

Disease 


Health 

Death 


Life 

Two-faced 


Single-faced 

Whimsical 


Constant 

Bad 


Good 

Demanding 


Giving 

Destructive 


Constructive 

(castrating) 


(virilizing) 

(mother) 


(father) 

(mother’s family) 


(father’s family) 

Possession 


Mysticism 

Jidba 


Hal (hizb/dikr) 

Dream (nightmare) 


Dream 

Bad luck 


Good luck 

Sacrifice 


Sacrifice 



226 


THE HAMADSHA 


To these the anthropologist, as an outsider, might well add to the list of 
‘Aisha’s associations “low tradition,” Dionysian, and expressed, and, to 
the saintly associations, “high tradition,” Apollonian, and repressed. 
It is clear that these two figures span several spheres of reality and are 
subject to a complex logic which must “fit” the demands of these dif¬ 
ference spheres 12 (Berger and Luckmann 1967:40). 

The figure of the saint, who is representative of the ideal social and 
moral order, is not altered in the attempted resolution of these polar 
opposites. He makes no special demands on the patient, although the 
patient may have promised him a sacrifice or some other gift. 13 Despite 
the saint’s ability to master the everyday affairs of the masculine world, 
as the legends attest, the role is too specialized, too removed from the 
daily concerns of the average patient, to provide a model with which 
he can identify. (This may in fact be indicative of the child’s image 
of the father.) Rather, the saint provides the patient with baraka, the 
symbolic equivalent of semen, virility, and the principle of patrilineal- 
ity, and it is this baraka which is held responsible for the patient’s cure. 
It revitalizes—“revirilizes” is perhaps more accurate—the patient and 
enables him to overcome his illness—which is, as we have seen, asso¬ 
ciated with femininity, his feelings of inadequacy and weakness, and 
his inability to live up to the standards of male behavior. It enables him 
to bring about the transformation of e Ai'sha Qandisha. 

Resolution of the opposition between the figures of saint and jinniyya 
is attempted by transforming ‘A'isha Qandisha. Although she retains all 
of her chthonian-Dionysian associations, she ceases to be a socially dis¬ 
ruptive force and, in fact, like the Eumenides, becomes a support of 
the moral and social order. 14 She no longer takes away the patient’s 
health, virility, fertility, or good fortune. Rather, she preserves them. 
The transformation is not simply a change in her character. It demands 
a change in the patient’s relationship—in his character—to her. The 
patient must first placate ‘Aisha and then follow her commands. 

“As a given of the world in which the individual is bom, they serve to organize 
his psychic reality; they become bnages of his experience which must be con¬ 
gruent with socio-cultural processes and perhaps even physiological ones. 

“The adepts of the urban zawiyas are required to work for the saint. ‘A'isha 
Qandisha docs not play as important a role for them. They must work for the 
saint in order to receive his baraka—to remain healthy, successful, and virile. 
They do not identify with the saint, but work for him as a son works for his 
father. Failure to work for the saint is conceived of as a loss of baraka and con¬ 
sequent vulnerability to attacks by c Aisha Qandisha and other jnun. The saint 
himself can do no harm. 

“That such a change is possible is suggested by C A is ha’s ability to appear as a 
beauty or a hag; however, she is usually considered to be more dangerous when 
she manifests herself as a beauty (Crapanzano 1971). 




THE EXPLANATION OF THERAPY 


227 


Placation involves ‘Aisha Qandisha’s taking possession of the patient 
and often forcing him to symbolically enact his own castration—that is, 
to slash his head. In order to be cured, the man must become a woman. 
Thanks to the baraka of the saint, he is able to pass through this femini¬ 
zation and become a man again. From a psychological point of view, he 
is provided with a structured and socially sanctioned opportunity to 
regress and to act, at least symbolically, on the unarticulated desire to 
become a woman. “There are many ways,” Bruno Bettelheim (1955: 
107) has written, “in which man can deal with a socially unacceptable 
desire. One way is to dramatize it, act it out, and through intense but 
in the long run only token satisfaction, try to rid himself of it perma¬ 
nently.” This “acting out” appears in the Hamadsha case to have at least 
some immediate therapeutic value. 

Having once placated the jinniyya, the patient must obey her com¬ 
mands, or she will attack him and make him ill. These commands— 
wearing certain colors, burning special incense, making periodic sacri¬ 
fices, and the like—seem arbitrary. I have found in the analysis of life 
histories that the breaking of a command is in fact, at one level symbolic 
of a breach in public or private morality and at another, symbolic 
of the feelings—the Westerner would call them “guilt”—which this 
breach has produced in the individual. The following example may help 
to clarify my point. 

Driss became a Hamdushi, he explained, after he laughed at the Hamadsha 
who were performing in the center of his village. He fell to the ground and 
was raised up by the spectators and forced to dance. He became a follower 
of 'Aisha Qandisha and followed her commands explicidy. Many years later 
—he was 14 at the time of the above incident—he again fell sick. “All my 
bones stung. My body was heavy. I could not move. My nose bled.” He 
did not know why he had fallen sick again. He had forgotten to buy some 
black jawi, he remembered, and had not burned any incense in a litde altar, 
which he had constructed for 'Aisha, for several days before falling sick. 
He continued to reminisce. His sister had been struck at the time and had 
wanted him and his wife to accompany them on a pilgrimage ( ziyara ). He 
had refused. He was sure that his sister’s husband had instructed her to ask 
him to go with her to Sidi 'Ali’s. He wanted to split the cost. (Driss, I had 
learned on another occasion, did not get on well with his sister’s husband, 
who somehow cheated him out of some money. Driss was not related to 
him by blood.) 

Driss’ case illustrates a dynamic that I have encountered in many 
Hamadsha life histories. The circumstances leading to the initial attack 
by the jinn were such that a Westerner, given the Moroccan Arab 



228 


THE HAMADSHA 


moral code, would describe them as having “caused a guilty con¬ 
science.” (In this case they were perhaps stereotyped and masked still 
another cause of guilt.) The very stance of the individual vis-a-vis him¬ 
self and others within his world—therapy itself—is here articulated in 
what Marie-Cecile and Edmond Ortigues, in their study of the Oedipus 
complex in Senegal (1966:196-197), have called “the novelistic mode 
of individual adventure and inner guilt,” but “in the tragic mode of 
persecution by the powers of destiny.” 15 Having once defined their 
“feelings” in terms of the jinn, the Hamadsha will interpret every sub¬ 
sequent “guilt reaction” of similar magnitude with reference to the jinn. 
Extreme reactions are often somatized in the form of paralysis or other 
conversion reactions. They are said to result from a failure to obey the 
jinn’s commands. These arbitrary commands become symbolic masks 
which may cover the actual moral fault (Crapanzano 1972). 

‘Ai'sha Qandisha is transformed from a force disruptive to the social 
and moral order into a force to preserve that order. So long as her fol¬ 
lower obeys her commands, he receives her support. So long as he 
follows his society’s moral code, she enables him to live up to the ideals 
of male dominance, superiority, and virility which are congruent with 
the extreme patrilineality and patripostestality of the Arab world. It is 
‘A'fsha Qandisha who keeps him a man, and not the saints. They provide 
only the baraka with which to transform her, and not a model with 
which to identify. Neither saint is represented as a strong, virile figure 
who has overcome the very ills his baraka is able to cure. 16 Sidi c Ali 
rendered himself a woman by playing the passive role before Bu'abid 
Sharqi and the other saints; Sidi Ahmed rendered himself a woman by 
symbolically castrating himself. 17 

If the follower fails to obey her commands, she attacks him and 


“ To be sure, at least in the cities, the Hamadsha cures—the jnun they work 
with—are not articulated to the lineage structure as they are in Senegal. The in¬ 
dividual Moroccan—and not his family or lineage—becomes dependent upon the 
jnun. No communal altar is established (Ortigues 1966:158). 

M It is noteworthy that the saints are also not figured as economically successful. 

17 The fact that both saints retain their baraka despite their feminization is 
suggestive of the roots of the Arab male’s attitude toward sex. Great emphasis 
is placed on the number of ejaculations, and not on foreplay or prolonged coitus. 
There are tales told, not without admiration, of men who ejaculated so many 
times in a night that their “vein,” which transforms blood into semen, was unable 
to keep up with them. They ejaculated blood. Such men were said to have had 
12 to 15 ejaculations in the night. The transformation of c Aisha by means of the 
baraka of the saint may mirror the man’s conquest and “domestication” of the 
woman by the sexual act. The woman, who is often foreign to the patrilineage, 
or at least the household, just as ‘Aisha Qandisha is foreign, coming as she does 
from the Sudan, is dangerous and must be transformed—domesticated. 



THE EXPLANATION OF THERAPY 


229 


makes him sick. He becomes as woman. ‘Ai'sha seems to function as a 
sort of externalized superego . 18 Her transformation reflects the master¬ 
ing of feelings of inadequacy, femininity, and remorse, resultant in part 
from the moral fault, and the conversion of these feelings into a socially 
and morally positive force. Mastery involves not a positive identifica¬ 
tion with a male figure but the individual’s regression to a state in which 
he is able to act out these feelings and, at the same time, to chastise 
himself through acts of self-mutilation or simply through membership 
in a cult looked down upon by other members of his society. 

The roots of these feelings of inadequacy and femininity must be 
sought in the lives of the Hamadsha. It is impossible here to determine 
whether they are rooted in a single trauma—such, for example, as cir¬ 
cumcision—or in a more complex behavioral syndrome, such as an 
authoritarian father, too arbitrary and aloof for the child to form a 
positive identification with him. In any case, the hadra does provide 
the individual with an occasion to discharge tensions associated with 
these feelings in a socially acceptable manner. The cures may result 
from anxiety reduction following the reenactment of past traumata, or 
the acting out of repressed impulses; or they may result from an actual 
restructuring of psychic reality—a change in personality. It may be 
the case that these elaborate symbolic transformations are epiphenome- 
nal and of little therapeutic consequence. The Hamadsha may indeed 
be compelled to relive the past’s experiences and to renew their rela¬ 
tionship with ‘Aisha Qandisha again and again in the hope of achieving 
permanent cure, and in fact obtain but temporary relief from the in¬ 
delible scars of their past. That the majority of Hamadsha patients 
become devotees of the brotherhood and must periodically perform the 
hadra, enter jidba, and on occasion slash their heads suggests that there 
is no permanency to the cure, that there is no personality change of 
long standing, and that the Hamadsha are destined to reenact their 
suffering. It is, after all, written. 

“Canscvcr in his study of the effects of circumcision on Turkish youth (1965) 
notes that the female is often perceived as the castrator. It should be pointed out 
that in the area from which the Hamadsha are recruited the mother leads her 
son to the house where he is to be circumcised. The father usually disappears 
for the occasion. The boy, who is usually under 5, is circumcised by a barber, 
usually a stranger to him, and held by an elder man of his village or family even. 
Immediately thereafter he is placed on his mother’s back and wrapped up in 
blankets. Warmth is said to aid the healing. His mother, who had been waiting 
in front of the door to the house, surrounded by villagers and musicians, begins 
to dance. Often she has pulled down the back of her dress “so that her clothes 
will not get all bloody,” the boy’s penis being pressed against her naked, sweating 
back. Men who remember the event recall with horror the sting of the salty 
sweat on their bleeding penises. 




Appendix 


Studies of trance behavior throughout the world have called attention 
to a number of factors that induce trance. I will consider here those 
factors which may play a role in the induction of trance among the 
Hamadsha. 

Sensory Deprivation and Sensory Overloading 

It has been suggested that trance and possession states, like other 
altered states of consciousness, “may be produced in any setting by any 
agents or maneuvers which interfere with the normal inflow of sensory 
or proprioceptive stimuli, the normal outflow of motor impulses, the 
normal ‘emotional tone,’ or the normal flow and organization of recog- 
nitive processes” (Ludwig 1968:70). Levels of exteroceptive stimula¬ 
tion, either above or below “an optimal level,” as well as an impover¬ 
ishment of varied and diversified environmental stimulation, appear 
conducive to the production of altered states of consciousness (Lud¬ 
wig 1968:70). Trance and possession states may be induced by either 
a reduction or an increase in exteroceptive stimulation. Sensory depri¬ 
vation is more commonly employed in laboratory studies and in hyp¬ 
nosis (Gill and Brenman 1966:4-11; 123-126), and sensory overloading 
is more commonly used in trancing ceremonies (Lee 1968:49). It has 
been further suggested that the effects on the organism of both sensory 
deprivation and sensory overloading are similar (Alland 1961:212). 

The Hamadsha employ primarily sensory overloading techniques to 
induce trance. Auditory overstimulation, which is produced by loud, 
monotonous drumming, the droning of the ghita, communal chanting, 
hand-clapping, and even the pounding of the male dancers in line, is 
the most important technique. Chanting is not as well developed among 
the Hamadsha as it is in other Islamic confraternities such as the Rahma- 
niyya of Algeria (Haas 1943). With the exception of darkening the 



232 


THE HAMADSHA 


room during the performance of ‘ATsha Qandisha’s rih, visual stimuli 
are not manipulated by the Hamadsha. Jallaba hoods and hair falling 
over the eyes, as well as the effects of rapid dancing, may, however, 
serve to reduce or alter the inflow of visual stimuli. The inhalation of 
incense can be considered as a means of producing olfactory over- 
stimulation. Alterations in gustatory and tactile stimuli—with the ex¬ 
ception of the dancing itself—are not of central importance in the 
Hamadsha ceremonies. Sensory deprivation—quiet and darkness—may 
play a role in the induction of trances in the saints’ tombs during a 
pilgrimage. 

Changes in Motor Activity 

Strict limitations of motor activity facilitate hypnotic trance 1 (Gill 
and Brenman 1966:6; 126-129), and may be of some significance in 
trances in the saints’ tombs. The Hamadsha encourage heightened 
motor activity in their dancing; the monotonous and stereotyped char¬ 
acter of these dances should not, however, be ignored. 

Ideational Deprivation and Thought Confusion 

Limitation in the range and variety of thought (Gill and Brenman 
1966:6), as well as techniques productive of thought confusion (Erik- 
son 1964), play an important role in hypnosis. Chanting, dancing, and 
music may also reduce “ideational stimuli.” Dance leaders, like the 
muqaddim who tried to exorcise the jinn in the 7-year-old girl, often 
attempt to confuse the patient. In hypnosis, ideational deprivation is 
associated with an attempt to alter bodily awareness. The Hamadsha 
do not make any active suggestions with regard to bodily awareness, 
but the phenomenology of the hal and jidba is well-articulated; it is 
well-known to all dancers and devotees. The expectations of the Ha¬ 
madsha may serve a function similar to that of the active suggestions 
of the hypnotist. 

Rhythmic Stimulation 

Attention has been called to the effect of drumming on ceremonial 
trance induction (Sargant 1964:92). Andrew Neher (1962:151-160) 
has noted the similarity between the effects of rhythmic flashing of 
light—photic drive—on laboratory subjects and the symptomatology of 

1 Although many authorities have attempted to equate hypnotic trance with 
ceremonial trance, I find this approach inadequate. It tends to lay emphasis on 
similarities and to blur what may be essential differences. This is not to say that 
induction techniques for both states may not be similar. 



APPENDIX 


233 


trancers. They include alterations in perception, especially visual, that 
are not present in the stimulus, kinaesthetic changes such as swaying, 
spinning, and vertigo, tingling and pricking sensations, fatigue, confu¬ 
sion, fear, disgust, anger, disturbances in time sense, hallucinations, 
“clinical psychopathic states,” and epileptoid seizures. A number of 
“precipitators” are known to aid simply rhythmic stimulation in pro¬ 
ducing these effects: rhythms that accompany the main rhythm, rhyth¬ 
mic stimulation of more than one sensory mode, stress, hyperventila¬ 
tion, low blood glucose and adrenalin production resulting from fatigue 
and over-exertion. Neher suggests that drumming has an effect on the 
individual which is similar to that of photic driving. 

Not only are the symptoms of photic driving similar to those re¬ 
ported by the Hamadsha, but many of the same “precipitators” are 
present. Several different drum rhythms are played at once; drumming 
is accompanied by dancing (kinaesthetic stimulation); many of the 
Hamadsha, especially the patients, are in stress; hypo- and hyper- 
ventilatory techniques are employed; prolonged dancing produces fa¬ 
tigue and in all probability a reduction in blood sugar and an increase 
in adrenalin production. The extent to which the rihs of the Hamadsha 
correspond to different brain rhythms and induce trance in those 
dancers whose brain rhythm “corresponds” to the rhythm of the par¬ 
ticular rih being played has yet to be explored. 

Special Breathing Techniques 

Breathing techniques which either over-oxygenate or under-oxygen¬ 
ate the blood are known to aid in the induction of trance (Haas 1943). 
Often they produce paresthetic sensations and visual changes that may 
be the result of an increase in carbon dioxide in the lungs and blood¬ 
stream (Alland 1961). Cerebral anoxia is a prominent and constant fac¬ 
tor in precipitating convulsive seizures (Harrison et al. 1950:97). Al¬ 
though the Hamadsha have not developed their breathing techniques 
to the same extent as orders such as the Rahmaniyya have, the dancers 
in line do hyperventilate. It is possible that members of the audience 
who do not join the line also hypo- and hyperventilate in time to the 
music or to the breathing of the dancers. The fact that one of the pro¬ 
dromal symptoms of trance is paresthesia would seem to suggest this. 

Drugs 

Critics of the Hamadsha and adepts of other brotherhoods sometimes 
claim that the Hamadsha’s trances are produced by kif (marihuana) 
and hashish, but I found no evidence for this whatsoever. In fact, few 



234 


THE HA MADS HA 


of the Hamadsha smoked kif at all, and even fewer smoked hashish. 
Although one or two other hallucinogenic drugs are employed in Mo¬ 
rocco—these include nutmeg—only two of my informants had taken 
them, and only on one or two occasions. 

Dancing 

Dancing can result in a state of fatigue and ovcrexcrtion and a conse¬ 
quent reduction of blood sugar and increase in adrenalin. Hypogly¬ 
cemia is known to produce convulsions, and a certain deteriorated form 
of adrenalin—androchrone—is thought to be hallucinogenic (Hoffer 
et al. 1954; Wallace 1956). 

Dietary Conditions 

The Hamadsha do not follow any dietary restrictions, such as pre- 
ceremonial fasting, which might aid in trance induction. In fact, before 
beginning to dance they often drink large quantities of highly sweet¬ 
ened tea. This would tend to raise the level of blood sugar and prevent, 
or at least postpone, a hypoglycemic seizure. Nevertheless, some of 
the seizures resemble those of insulin shock. This is true also of Balinese 
trance seizures (Mead, Margaret, personal communication). It is pos¬ 
sible that some of the patients suffer a loss of appetite, but this is not 
a typical symptom of the Hamadsha patients. Certainly, their diets 
merit a detailed examination. Consideration should be given to the rela¬ 
tionship between a low protein diet and calcium metabolism, since hy¬ 
pocalcemia is known to produce convulsions (Wallace 1961). 

Heat and Stuffiness 

Although heat and stuffiness with a consequent increase in atmos¬ 
pheric carbon dioxide have been considered factors in the induction of 
trance (Alland 1961), neither factor appears to be of great significance 
in the case of the Hamadsha. Ceremonies are often held out of doors, 
and even those which take place inside are well ventilated. 2 

Ambiance 

The ambiance of the ceremonies plays an important part in trance 
induction. The following factors appear to be significant: (1) Group 
excitement and overcrowding. (It should be noted that Hamadsha 
ceremonies are among the few occasions in which men and women are 
not strictly segregated.) (2) Heightened expectation. (3) Theatricality 

8 Women’s quarters do tend to be stuffy, however. 




APPENDIX 


235 


and violence. (4) A permissive atmosphere. (5) Presence of strong 
models for trance and mutilation. 

Leadership 

One of the most significant factors in the induction of hypnotic 
trance is strong, authoritative leadership. Gill and Brenman (1966:xix- 
xx) have defined hypnosis as “a particular kind of regressive process 
which may be initiated by sensorimotor-ideational deprivation or by 
the stimulation of an archaic relationship to the hypnotist.” They found 
that when one of these techniques was used, phenomena characteristic 
of the other began to appear. Investigators who have been anxious to 
equate the ecstatic experiences of the Sufi mystic with hypnotic trance 
have emphasized the role of the Sufi master and have considered his 
relationship to the adepts as analogous to the transference relationship 
between patient and psychotherapist (Shafii 1968). Although the role 
of the master is of paramount importance in many Islamic brother¬ 
hoods, the muqaddim of a Hamadsha team plays an unimportant role 
in the lives of the adepts and devotees during the ceremonies. 

Individual Factors 

There are probably a number of individual factors which also help 
to induce trance. These would include physiological ones such as sugar 
metabolism and psychological ones such as stress, anxiety, and emotional 
deprivation. 



Glossary 


e afarit: especially powerful and terrifying fnun (spirits). 
afyival: see hal. 

*ar: lit., shame; a vow or promise. 
ary ah: see rib. 

baraka : holiness, blessing, good fortune; a spiritual or miraculous force 
that emanates from holy men and places. Bread is a recurrent sym¬ 
bol for baraka. Baraka can be either inherited or achieved. 
bidonville: a shantytown. Such shantytowns have grown up around 
Meknes and other Moroccan cities since the arrival of the French. 
burban: manifest evidence, proof; often used as a synonym for “mira¬ 
cle” ( karama ). 

debiha: an animal sacrifice. 

dirham: Moroccan money, worth about 20 cents in American money 
at the time of this study. There are 100 Moroccan francs to a 
dirham 

dikr: a short, rapid invocation. 
fatha: a short prayer 

foqra (sing., fqir ): members of a zawiya; adepts. 
fqih: a Koranic teacher. 

ganbri: a three-stringed guitar, often made from a turtleshell. 
ghita: an oboe, played by ghiyyata (sing., ghiyyat). Smaller ones are 
used in marriage and circumcision ceremonies; larger ones— ghita 
gharbaiviyya —are used in the Hamadsha hadra. 
ghwal (m. sing., ghul; f. sing., ghula ): ogres, ghouls. 
gwal: an hourglass-shaped pottery drum, played by guu^wala (sing., 
guimval). 



238 


THE HAMADSHA 


hadiyya: gift. 

hadra: the ecstatic dance of the Hamadsha and other religious brother¬ 
hoods. 

hadra gnawiyya : the last part of the hadra, in which musical phrases 
allegedly from the Gnawa are played. 
fyal (pi., ah'wal ): state, circumstance, time; an ecstatic trance. 

Hamadsha (m. sing., Hamdushi ; f. sing. Hamdushiyya): Members of 
the loosely and diversely organized religious brotherhood known as 
at-tariqa al-hamdushiyya. Followers of Sidi c Ali ben Hamdush 
( c Allaliyyiii) and Sidi Ahmed Dghughi (Dghughiyyin) arc collec¬ 
tively referred to as Hamadsha. 

hizb: a chant or litany, usually containing verses of the Koran. 
hufra : grotto, cave, pit. 
horm: a sacred place. 

jawi: incense; benzoin. 

jidba: ecstasy; a frenetic trance state. 

jnun (m. sing., jinn; f. sing., jinmyya ): devils, imps; spirits; sprites. 

khalifa-, assistant; assistant to the leader ( muqaddim) of a zaiviya. 

lila: lit., night; used to refer to a Hamadsha ceremony ( sadaqa ) which 
takes place in the evening and in which the hadra is performed. 

madina: town, city; specifically, in the text, the old quarter of Meknes. 
marabout: see murabit. 

mizwar: the leader of a lineage group descended from a saint. 
msamiad : supported, sustained by; used in the Hamadsha context with 
reference to a special relationship one has with a saint. 
muhibbin (m. sing, muhibb): devotees. 

muqaddmi: the caretaker of a saint’s tomb ( qubba)\ the leader of a 
Hamadsha lodge or team. 

murabit: a man attached to God. The French word viaraboutisme , a 
catchall word for activities associated with the worship of Muslim 
saints, is derived from it. 
musem: an annual pilgrimage to a saint’s tomb. 

muttakil and muwali: both words mean leaning on, or near; used in 
the Hamadsha context with reference to a special relationship one 
has with ‘AYsha Qandisha and other jnun. 

nira: a recorder, made from a reed. 

qubba: a saint’s tomb, usually a squat, white, cubical building with a 
domed roof. 




GLOSSARY 


239 


rih (pi., ary ah): lit., air, wind; a musical phrase, played during a 
hadra, which is said to be the favorite of a particular jinn and sends 
a follower into trance. 

sadaqa : charity, alms; also a Hamadsha ceremony in which the hadra 
is danced and a communal meal is served. 
sharif (pi., shurfa ): a descendant of the Prophet Muhammad. 
shayatin: evil spirits, children of the devil ( Shitan ). 
siyyid: a saint; the tomb of a saint. 

tabbal: one who plays the snare drum ( tabil ). 
tab?: the follower of a saint or jinn. 

tdifa: a team; used specifically in the text to refer to a team of bidon- 
ville Hamadsha, although it may in fact also be used to refer to the 
members of a zawiya. 

talla e : an exorcist-seer who identifies attacking jnun. 
tariqa: lit., the way, path; a spiritual path, the teachings of a particular 
saint. Also used to refer to the religious brotherhoods. 

'wali: a saint. 

•wulad siyyid: lit., “the children of the saint”; the agnatic descendants 
of a saint. 

zawiya: lit., a comer; the lodge or meeting place of a religious broth¬ 
erhood. Used in the text to refer to both the meeting place and the 
members of an urban Hamadsha team. 
ziyara: visit; pilgrimage to a saint’s tomb. Also used to refer to offer¬ 
ings left by a pilgrim at a saint’s tomb. 



Bibliography 


Abun-Nasr, Jamil M. 

1965 The Tijaniyya: a sufi order in the modern world. London: Ox¬ 
ford University Press. 

Alland, Alexander 

1961 “Possession” in a revivalistic Negro church. Journal for the Scien¬ 
tific Study of Religion 1:204-213. 

Andrews, J.B. 

1903 Le culte des fontaines de Sebba Aioun a Alger. Alger: Jourdan. 

Anonymous 

1927? Les zaouias marocaines et le mekhzen (Manuscript 31 H 1). Aix-en- 
Provence: Depot des Archives d’Outre-Mer. 

Aubin, Eugene 

1904 Le Maroc d’aujourd’hui. Paris: Armand Colin. 

Barnes, J. A. 

1968 Networks and political process. In Local-level politics: social and 
cultural perspectives, M. M. Swartz, ed., Chicago: Aldine Publish¬ 
ing Company, pp. 107-130. 

Basset, Henri 

1920 Le culte des grottes au Maroc. Alger: Ancienne Maison Bastide- 
Jourdan, Jules Carbonel. 

Bastide, Roger 

1958 Le condomble de Bahia. Paris and The Hague: Mouton. 

Belo, Jane 

1960 Trance in Bali. New York: Columbia University Press. 

Berger, Peter L., and Thomas Luckmann 

1967 The social construction of reality: a treatise in the sociology of 
knowledge. Garden City, New York: Anchor Books. 

Berque, Jacques 

1958 Medinas, villeneuves, et bidonvilles. Les Cahiers de Tunisie 6:5-42. 

Bettelheim, Bruno 

1955 Symbolic wounds: puberty rites and the envious male. London: 
TTiamcs and Hudson. 



242 


THE HAMADSHA 


Boulanger, Robert 

1966 Morocco. Paris: Hachette. 

Bourgignon, Erika 

1968 Divination, transe et possession en Afrique transsaharienne. In La 
divination, Vol. 2, Andre Caquot and Marcel Leibovici, eds., Paris: 
Presses Universitaires de France, pp. 331-358. 

Brignon, Jean, A. Amine, B. Boutaleb, G. Martinet, and B. Rosenberger 

1967 Histoire du Maroc. Paris: Hatier. 

Brunei, Rene 

1926 Essai sur la confrerie religieuse des c Aissaoua au Maroc. Paris: Paul 
Geuthner. 

1955 Le monarchisme errant dans l’lslam: Sidi Heddi et les Heddawa. 
Paris: Larose. 

Canal, Joseph 

1902 Geographic generale du Maroc. Paris: Augustin Challamel. 

Cansever, Gocke 

1965 Psychological effects of circumcision. British Journal of Medical 
Psychology 38:321-331. 

Casablanca et les Chaouia 

1915 In Villes et tribus du Maroc, Vol. 1, Paris: Leroux. 

Cat, Edouard 

1898 L’lslamisme et les confrdries religieuscs au Maroc. Revue des Deux 
Mondes 68:375-404. 

Chclhod, Joseph 

1964 Les structures du sacr6 chez les Arabes. Paris: Maisonneuve et 
Larose. 

Chottin, Alexis 

1938 Tableau de la musique marocaine. Paris: Geuthner. 

Coon, Carleton S. 

1964 Caravan: the story of the Middle East, rev. ed. New York: Holt, 
Rinehart, and Winston. 

Crapanzano, Vincent 

1971 The transformation of the Eumenides: a Maroccan example. Paper 
read at the annual meeting of the American Anthropological Asso¬ 
ciation. 

1972 The Hamadsha. In Saints, scholars and sufis, Nikki Keddie, ed., 
Berkeley: University of California Press. 

Crapanzano, Vincent, and J. Kramer 

1969 A world of saints and she-demons. The New York Times Maga¬ 
zine (June 22): 14-38. 

Delehaye, Hippolyte 

1955 Les legendes hagiographiques. Bruxelles: Societe des Bollandistes. 

Depont, O., and J. Coppolani 

1897 Les confreries religieuses musulmanes. Alger: Adolphe Jourdan. 



BIBLIOGRAPHY 


243 


Dermenghem, Emile 

1954 Le culte des saints dans l’lslam maghrebin. Paris: Gallimard. 

Dodds, E. R. 

1966 The Greeks and the irrational. Berkeley: University of California 
Press. 

Dundcs, Alan, Jerry W. Leach, and Bora Ozkok 

1970 The strategy of Turkish boys’ verbal dueling rhymes. Journal of 
American Folklore 83:325-349. 

Draques, Georges 

n.d. Esquisse d’histoire religieuse du Maroc: confreries et zaouias. Paris: 
Peyronnet. 

Eliade, Mircea 

1965 Le sacrd et le profane. Paris: Gallimard. 

Erikson, Milton 

1964 The confusion technique in hypnosis. American Journal of Clinical 
Hypnosis 6:183-207. 

Evans-Pritchard, E. E. 

1949 The Senusi of Cyrenaica. Oxford: Clarendon Press. 

Fenichel, Otto 

1966 The psychoanalytic theory of neurosis. London: Routledge and 
Kegan Paul. 

Festinger, L., H. W. Riecken, and S. Schachter 

1956 When prophecy fails. New York: Harper and Row. 

Figueras, Tomas Garcia 

n.d. Hamadcha y Aisaua (manuscript). Madrid: Biblioteca nacional. 
Fortes, Meyer 

1953 The structure of unilineal descent groups. American Anthropolo¬ 
gist 55:17-41. 

Franchi, Jean 

1959 Urbanisation d’un bidonville: Bordj Moulay Omar (Meknes). Bul¬ 
letin Economique et Social du Maroc 23:255-291. 

Frazer, James George 

1961 Adonis, Attis, Osiris: studies in the history of Oriental religion. 
New Hyde Park, New York: University Books. 

Freud, Sigmund 

1941 Medusa’s head. International Journal of Psychoanalysis 22:69. 

1963 A neurosis of demonical possession. In Freud: studies in parapsy¬ 
chology, P. Rieff, ed., New York: Collier Books, pp. 91-123. 
Garcia-Barriuso, P. P. 

1941 La musica hispano-musulmana en Marruecos. Larache: Artes- 
Graficas Bosca. 

Gardet, L. 

1960 Al-burhan. In Encyclopaedia of Islam, New Edition, Vol. 1, pp. 
1326-1327. 



244 


THE HA MADS HA 


Geertz, Clifford 

1968 Islam observed: religious development in Morocco and Indonesia. 
New Haven: Yale University Press. 

Gellner, Ernest 

1963 Saints of the Atlas. In Mediterranean countrymen, J. Pitt-Rivers, 
ed., Paris: Mouton, pp. 145-157. 

1969 Saints of the Atlas. London: Weidenfeld and Nicolson. 

Gerth, H. H., and C. W. Mills 

1958 From Max Weber: essays in sociology. New York: Oxford Uni¬ 
versity Press. 

Gibb, H. A. R. 

1961 Mohammedanism: an historical survey, 2nd ed. London: Oxford 
University Press. 

Gibb, H. A. R., and H. Bowen 

1957 Islamic society and the west: a study of the impact of western 
civilization on Moslem culture in the Near East, Vol. 1, part 2. 
London: Oxford University Press. 

Gill, M. M., and M. Brenman 

1966 Hypnosis and related states: psychoanalytic studies in regression. 
New York: Science Editions. 

Goldziher, Ignace 

1880 Le cultc des saints chez les Musulmans. Revue de l’Histoire des 
Religions 2:257-351. 

Greenberg, Joseph 

1946 The influence of Islam on a Sudanese religion. Monographs of the 
American Ethnological Society 10: vii—73. 

Haas, William S. 

1943 The zikr of the Rahmaniya order: a psychophysiological analysis. 
The Moslem World 33:16-28. 

Haimer, Mehdi 

1968 Aicha Kandicha. Lamalif 28:52. 

Harrison, R. R., H. H. Merritt, and D. Sciarra 

1950 Coma, convulsions, and paralysis. In Principles of internal medicine, 
T. R. Harrison, ed., Philadelphia: Blackiston, pp. 91-109. 

Herber, J. 

1923 Les Hamadcha et les Dghoughiyyin. Hesperis 3:217-235. 

Hoffer, A., H. Osmond, and J. Smythies 
1954 Schizophrenia: a new approach, Part II: result of a year’s research. 
Journal of Mental Science 100:29-45. 

Howell, David 

1970 Health rituals at a Lebanese shrine. Middle East Studies 6:179-188. 
Jeanmaire, H. 

1949 Le traitement et la mania dans les “mysteres” de Dionysos et des 
Corybantes. Journal de Psychologic 42:64-82. 

1951 Dionysos: histoire du culte de Bacchus. Paris: Payot. 



BIBLIOGRAPHY 


245 


Jones, E. 

1951 The madonna’s conception through the ear: a contribution to the 
relation between aesthetics and religion. In Essays in applied psy¬ 
choanalysis, Vol. 2, London: Hogarth Press, pp. 266-357. 

Julien, C. A. 

1966 Histoire de l’Afrique du Nord, Vol. 2. Paris: Payot. 

Lane, Edward 

1963 The manners and customs of the modem Egyptians. London: 
Everyman’s Library. 

Lecoeur, C. 

1969 Le rite et l’outil:essai sur le rationalisme social et la plurality des 
civilisations, 2® ed. Paris: Presses Universitaires de France. 

Lee, Richard 

1968 The sociology of the !Kung Bushman trance performance. In 
Trance and possession states, R. Prince, ed., Montreal: R. M. Bucke 
Memorial Society, pp. 35-54. 

Leiris, Michel 

1958 La possession et ses aspects theatraux chez les Ethiopiens de Gondar. 
L’Homme 1. 

LeToumeau, Roger 

1949 Fez avant le protectorat: etude economique et sociale d’une ville de 
l’occident musulman. Casablanca: Publications de l’lnstitut des 
Hautes Etudes Marocaines 45:1-668. 

1958 La decadence Sa’adienne et l’anarchie marocaine au XVII e siecle. 
Annales de la Faculte des Lettres d’Aix 33:187-225. 

Levi-Strauss, Claude 

1963a. The effectiveness of symbols. In Structural anthropology. New 
York: Basic Books, pp. 186-205. 

1963b The structural study of myth. In Structural anthropology, New 
York: Basic Books, pp. 206-231. 

Lewis, I. M. 

1969 Spirit possession in northern Somaliland. In Spirit mediumship and 
society in Africa, John Beattie and John Middleton, eds., London: 
Routledge and Kegan Paul, pp. 188-219. 

Lienhardt, Godfrey 

1967 Divinity and experience: the religion of the Dinka. Oxford: Claren¬ 
don Press. 

Ludwig, Arnold M. 

1968 Altered states of consciousness. In Trance and possession states, 
R. Prince, ed., Montreal: R. M. Bucke Memorial Society, pp. 
69-95. 

Margoliouth, D. S. 

1934 Shadhiliya. In Encyclopaedia of Islam, Vol. 4, pp. 247-249. 

Massignon, Louis 

1934 Tasawwuf. In Encyclopaedia of Islam, Vol. 4, pp. 681-685. 



246 


THE HAMADSHA 


Mercicr, Henry 

1951 Dictionnaire Arabe-Frangais. Rabat: Editions La Porte. 

Mercier, L. 

1906 Les mosquees et la vie religieuse a Rabat. Archives marocaines 
8:99-195. 

Metraux, Alfred 

1959 Voodoo in Haiti. New York: Oxford University Press. 

Michaux-Bellaires, E. 

1913 Le Gharb. Archives marocaines 20:1-477. 

1921 Essai sur l’histoire des conferies marocaines. Hesperis 1:141-159. 

1927 Conferences. Archives marocaines 27:1-334. 

Michaux-Bellaires, E., and George Salmon 

1906 Let tribus arabes de la vallee du Le Khous, 3ieme partie. Archives 
marocaines 6:219-397. 

Miner, Horace, and George DeVos 

1950 Oasis and casbah: Algerian culture and personality in change. Ann 
Arbor: Anthropological Papers, Museum of Anthropology, Uni¬ 
versity of Michigan, XV. 

Mischel, Walter and Frances 

1958 Psychological aspects of spirit possession. American Anthropologist 
60:248-260. 

Mole, Marijan 

1963 La danse extatique en Islam. In Les danses sacrees, Paris: Editions 
du Seuil. 

Montet, Edouard 

1902 Les confreries religieuses de l’lslam marocain: leur role religieux, 
politique et social. Revue de l’Histoire des Religions 45:1-35. 

Neher, Andrew 

1962 A physiological explanation of unusual behavior in ceremonies in¬ 
volving drums. Human Biology 34:151-160. 

Nicholson, Reynold A. 

1963 The mystics of Islam. London: Routledge and Kegan Paul. 

Ortigues, Marie-Cccile and Edmond 

1966 Oedipe africain. Paris: Libraire Plon. 

Paques, Viviana 

1964 L’arbre cosmique dans la pensee populaire et dans la vie quotidienne 
du nord-ouest africain. Paris: Travaux et memoires, Institut d’Eth- 
nologie 70:i—702. 

Paquignon, Paul 

1911 Notes et documents. Revue du Monde Musulman 14:525-536. 

Parsons, Talcott 

1964 Social structure and personality. New York: The Free Press. 

1966 Societies: evolutionary and comparative perspectives. Englewood 
Cliffs, N.J.: Prentice-Hall. 




BIBLIOGRAPHY 


247 


Peters, Enirys 

1960 The proliferation of segments in the lineage of the Bedouin of 
Cyrenaica. Journal of the Royal Anthropological Institute 90:29-53. 

Planhol, Xavier de 

1959 The world of Islam. Ithaca, New York: Cornell University Press. 

Prince, Raymond 

1968 Can the EEG be used in the study of possession states? In Trance 
and possession states, R. Prince, ed., Montreal: J. M. Bucke Memo¬ 
rial Society, pp. 121-137. 

Quedenfeldt, M. 

1886 Aberglaube und halbrcligiose brudcrschaftcn bei den Morokkancrn. 
Zcitschrift fur Ethnologie 28:671-692. 

Radcliffe-Brown, A. R. 

1962 Introduction. In African systems of kinship and marriage, A. R. 
Radcliffe-Brown and D. Forde, eds., London: Oxford University 
Press, pp. 1-85. 

Rahman, Fazlur 

1968 Islam. Garden City, New York: Anchor Books. 

Rinn, Louis 

1884 Marabouts ct khouan: etude sur l’lslam en Algerie. Alger: Adolphe 
Jourdan. 

Ritter, Helmut 

1955 Das meer dcr scelc: mcnsch, welt, und gott in den geschichten des 
Fariduddin 'Attor. Leiden: Brill. 

Salmon, G. 

1904 Confreries et zaouyas dc Tanger. Archives marocaincs 2:100—114. 

Sargant, William 

1964 Battle for the mind. London: Pan Books. 

Schoen, Captain 

1937 Les confreries musulmans dans le sud marocain. Report 89 of the 
Centre des Hautes Etudes d’Administration Musulmane, Paris (un¬ 
published). 

Service d’Urbanisme 

n.d. Meknes. Rabat: Ministere des trauvaux publics. 

Shafii, Mohammed 

1968 The pir (Sufi guide) and the western psychotherapist. Newsletter- 
Review of the R. M. Bucke Memorial Society 3:9-19. 

Spiro, Melford 

1967 Burmese supernaturalism: a study in the explanation and reduction 
of suffering. Englewood Cliffs, New Jersey: Prentice-Hall. 

Talha, Abdelouahed ben 

1965 Moulay Idriss du Zerhoun: quelques aspects de la vie sociale et 
familiale. Rabat: Editions Techniques Nord-Africaines. 



248 


THE HAMADSHA 


Terasse, Henri 

1930 Histoire du Maroc des origines a 1’etablissement du Protectorat 
frangais, 2 vols. Casablanca: Editions Atlantides. 

Tremeame, Arthur J. N. 

1914 The ban of the bori: demons and demon-dancing in West and 
North Africa. London: Health, Cranton and Ouseley. 

Turner, Victor W. 

1967 The forest of symbols: aspects of Ndembu ritual. Ithaca, New 
York: Cornell University Press. 

1969 The ritual process: structure and anti-structure. Chicago: Aldine 
Publishing Company. 

Van Grunebaum, G. E. 

1954 Islamic studies and cultural research. In Studies in Islamic cultural 
history, G. E. Van Grunebaum, ed., Memoir of the American An¬ 
thropological Association 76:1-22. 

1955 The structure of the Muslim town. In Islam: essays in the nature 
and growth of a cultural tradition, Memoir of the American An¬ 
thropological Association 81:141-158. 

Voinot, L. 

1948 Pelerinages judeo-musulmans du Maroc. Paris: Larose. 

Wallace, Anthony, F. C 

1956 Mazeway resynthesis: a bio-cultural theory of religious inspiration. 
Transactions of the New York Academy of Sciences 18:626-638. 

1959 Cultural determinants of response to hallucinatory experience. 
A.M.A. Archives of General Medicine 1:74-85. 

1961 Mental illness, biology, and culture. In Psychological anthropology: 
approaches to culture and personality, Francis L. K. Hsu, ed., 
Homewood, Illinois: Dorsey Press, pp. 255-295. 

Weigert-Vowinkel, Edith 

1938 The cult and mythology of the Magna Mater from the standpoint 
of psychoanalysis. Psychiatry 1:347-378. 

Westermarck, Edward 

1926 Ritual and belief in Morocco, Vols. 1 and 2. London: Macmillan. 

Windus, John 

1725 A journey to Mequinez. London: Jacob Jonson. 

Wittkower, E. D. 

1971 Transcultural psychiatry. In Modem perspectives in world psy¬ 
chiatry, J. G. Howells, ed., New York: Brunner-Mazel, pp. 697- 
712. 

Yalman, Nur 

1964 The structure of Sinhalese healing rituals. In Religion in South 
Asia, Edward B. Harper, ed., Seattle: University of Washington 
Press, pp. 115-150. 



Index 


‘Abdelhaqq, 32. 40. 41 
Abel, Theodora, 202 
Abreaction, 219-234 
Absinthe, in cures, 129 
Abu Medyan, 18,19.2Q 
Adepts ( foqra ): definition of, 75; 
and zawiyas, 80, 88; relationship to 
saints, 83, 186,226. 226n; social po¬ 
sition of, 85; families of, 88* See 
also Foqra 

£ Afarit: description of, 135. 136, 112* 
See also Jnun 

'Afrita, c Aisha Qandisha as, 44 
Aggression, during trance, 200 
Aguza, 111 
And, 164 
Ahidus, 191 

e Aisha Qandisha, xiii; description of, 
43, 44, 143, 145-146: legends of, 
43, 142-146: grotto of, 65. 170. 171. 
175, 182: shrines to, 80,91, 95. 116. 
143-144: and zawiya, 80, 98; and 
cures, 138. 141. 225, 227: marriage 
to, 143-145: commands of, 145, 
228: attacks by. 150. 160. 178. 183. 
187, 224, 226n; in hadra, 145, 196. 
201-202. 204; possession by, 164— 
165: and pilgrimages, 169. 171. 123; 
dreams of, 174; gifts to, 125, 122; 
sacrifices to, 184; as wasp, 201; 
transformation of, 226. 229; as ex¬ 
ternalized superego, 229. See also 
Jnun 

'Allaliyyin, 3; definition of, 75; role 
of mizwar, 80; professions of, 86; 


ethnic background of, 82; progres¬ 
sive position of, 182, See also 
Dghughiyyin; Hamadsha 
Alms, 109. See also Sadaqa 
Altered states of consciousness, 21L 
See also Hal; Jidba; Trance 
Amulets: at e Aisha Qandisha’s grotto, 
176; disposal of, 181 
Androgyny: and passage of baraka, 
53. 54; and male envy of female, 
55; unsatisfactory, 56. See also 
Feminization 

Angels: in legendary cycle, 41; con¬ 
fused with jnun, 136 
Anxiety reaction, 5,214 
c Ar: and c AIsha Qandisha, 80; defini¬ 
tion of, 159, 166; made by, 176; as 
sacrifice, 172; an d Sidi ‘Ali, 182. 
See also Debiha; Dhiya; Sacrifice 
Arabs, 7-8: Hilalian, 26 
1-Arusi, Sidi Ahmed: 46 
Asylum, in saints’ tombs, 22 
‘Awwad, 34-43 

Axe ( shaqur ), 46-47: at musem, 116. 
2£U 

‘Ayn Kabir: in legends, 32. 34. 37. 
38, 43; in pilgrimage, 65, 171. 172. 
175: and baraka, 65, 166 

Baraka: definition of, 2, 18; curative 
powers, 4, 11, 166. 167: and saints, 
31; passage of, 36, 32, 45, 48, 118- 
125. 135. 167-168. 187. 226n; and 
women, 48-49: personal, 52, 55. 
120: institutional, 52, 55, 120; and 












250 


NDEX 


feminization, 52, 228; as symbol, 
39, 53, 120, 226; in waters of e Ayn 
Kabir, 63_j endowment of, 73—74; 
and shantytown teams, 107; logic 
of, 119-120; and trance, 121; in 
pilgrimages 170, 174] in tuba, 174: 
social significance, 219. See also 
Saints 

Baraka-wealth exchange, 118-125 
Barbers, Ill 

Basket, in ceremonies, 188 
Bath: in tomb of Sidi c Ali, 68] in 
ceremonies, 188 

Beggars, and Hamadsha shrines. 111 
Beni Jennad, 64 
Beni Mtir, 148 

Beni Ouarad, 3] location of tomb of 
Sidi Ahmed, 45j description, 63- 
66; social organization, 66-71: pil¬ 
grimage to, 26 

Beni Rachid, 3; location of tomb of 
Sidi e Ali, 37j description, 63-66; 
social organization, 66-71: pilgrim¬ 
age to, 96; baths of, 116 
Ben join (jazvi), 138. See also Jawi 
Be rechid, 111 
Berger, Peter, 5-6, 7* 213 
Berque, Jacques, 104 
Bettelheim, Bruno, £L 221 
Bidonville. See Shantytowns 
Birth, theories of, 48-49, 49n 
Blacks, characteristics associated 
with, 85-86 
Blad 1-Makhzan, 62 
Blessing, cost of at shrine, 123 
Blindness, 4 

Blood: and baraka, xiv, 50] and 
e ATsha Qandisha, 145; effect on 
possessed, 162 

Bordj Moulay Omar, 105. 106 
Boualem, 91-96 

Bread: in miracles, 37] during musem, 
116; in wealth exchange, 118 
Breasts, mutilation of, 198n 
Brotherhoods, 2-3. 17-18; 20; in 
Meknes, 78-79, in Shantytowns, 
106-107. See also individual entries 
Bukhara, 8Z 


Burhan. See Miracles 
Buttermilk, in legends, 40,41 

Camel: in legends, 39j imitation of, 
201 

Candles, 173. 174, 126 
Canonists {^ilama), 18, in legends, 32 
Castration: and passage of baraka, 
52; and headslashing, 53] as placa- 
tion of c AIsha Qandisha, 227; and 
Sidi Ahmed, 228: and women, 229n 
Catafalque (darbuz)i64 
Cataleptic state, during trance, 122 
Catharsis. See Abreaction 
Ceremonies, xii-xiv; at zawiya, 97- 
98; by shantytown teams, 107, 210- 
211; financing, 107, 210-211: com¬ 
parison of zawiya and team, 186- 
187, 210; and passage of baraka, 
187; preparations for, 187-190: 
paraphernalia for, 188] litanies in, 
190-192; end of, 209-210: group 
support in, 215; cathartic function 
of, 216. See also Hadra; Lila; Sa- 
daqa 

Chanting, during hadra. See Litanies 
Charity. See Sadaqa 
Circumcision: invitations to, 74] and 
attacks by jnun, 140j ritual of, 229n 
Conflict, articulation and resolution 
of, 222 

Couscous, 99,116,118 
Cowrie shells, in divination, 162 
Crisis reaction, 214 
Cult membership: and cures, 5, 218- 
219; and change in social status, 5, 
219. See also Maraboutism 
Cures, 4, explanation of, 166-168; 
during pilgrimages, 169, 170; by 
temporary feminization, 227. See 
also Therapy 

Daff. 34-43 

Dance: description of, xii-xiv, 192— 
204, 207-210; in Sufism, 17] fre¬ 
quency of, 96] significance of, 141; 
and baraka, 186; women in, 196- 
199; men in, 199-204; and trance 
















INDEX 


251 


induction, 199-204. See also Hadra; 
Musem 

Darbala, 34,41 
Darqawiyya, 22 
Dates, as sign of welcome, 1S2 
Debiha, 177. See also e Ar; Dhiya; 
Sacrifice 

Delchaye, Hippolyte, UQ 
Depression, 4, 200n 
Devotees ( muhibbm ), 75i in madina, 
89-97: in Shantytowns, 107, 111- 
113; organization of 115, 125=128. 
See also Muhibbin 

Dghughiyyin, 3, 24] definition of, 75: 
professions of, 86] ethnic back¬ 
ground of, 87] conservative posi¬ 
tion of, 187: hizb of, 120 
Dhiya, LZ2* See also e Ar; Debiha; 
Sacrifice 

Diagnostics, 149-150. 153-156; in 
symbiotic cures, 159-165 
Dikr, 17] definition of, 186, 191; in 
madina ceremonies, 191-192; in 
shantytown ceremonies, 211. See 
also Hizb; Litanies; Wanasa 
Dionysius, 78, 22 

Divination: and diagnostics, 162-165; 
directed free association in, 163. 
See also Diagnostics; Exorcist seer 
Dominance, theme in legends, 48 
Drawa, 86, 82 

Dreams: and saints, 84, 170, 174, 178; 
and jnun, 139. 160; as goal of pil¬ 
grimage, 120; and cures, 180 
Drumming, in trance induction, 212 
Drums, in ceremonies, 188, See also 
Daff; Gwal; Tabil 
Dyafa, li2 

Ecstasy (wajd)^ LZ 
Ejaculation: during trance, 200: in 
sexual intercourse, 228n 
Epilepsy, and Hamadsha cures, 5 
Etiology, 134-135 

Eumenides, £ AIsha Qandisha com¬ 
pared to, 226 
Evil eye, 142 

Exorcism: and fuqaha, 135: and ther¬ 


apeutic theory, 157-159. See also 
Symbiotic cures 

Exorcist-seer: in Hamadsha cures, 
162-165; and pilgrimages, 121 
Explicative mode of responsibility, 
150-151 

Fatha: definition of xii, 189n; in elec¬ 
tion of muqaddim, 80] by mizwar, 
119; in shantytown ceremonies, 211 
Father: position of family, 8, 9] rela¬ 
tionship to son, 224 
Feminization: in legends, 51-52. 56; 
illness as, 224.226; and curing, 227; 
and baraka, 228; of saints, 228n 
Fig tree: at zawiya, 80] among Chao- 
uia, 91] at e Aasha Qandisha’s grotto, 
176.122 
Filala, 86, 82 

Flag, in madina ceremonies, 188 
Followers ( tabi c ): relation to jnun 
84-85. 142, See also Tabi e 
Foqra: definition of, 4] general de¬ 
scription of, 83-89: within Hamad¬ 
sha organization, 115; as poor, 118- 
119; as transfer agents of baraka, 
121; attitude toward mizwar, 124; 
position in city, 127. See also 
Adepts 
Fraja, 40 

Frogs, in legend, 40,44 
Fuqaha: definition of, 132; control 
over jnun, 135, U2, See also Ko¬ 
ranic teacher 
Futuhat, 61 

Ganbri, 89 

Geertz, Clifford, 19-20.119 
Genealogies: of foqra, 8] of wulad 
siyyid, 8, 86-71; Boualem, 91-92 
Ghaib (absence), 154-155 
Gharb, 82 
al-Ghazzali, 16,18 
al-Ghezwani, 21 

Ghita, xi; in hadra, 88, 182] relation 
to hal f 88 

Ghiyyata, xi, 75] in shantytown 

















252 


INDEX 


teams, 108-110; in therapeutic 
practices, 167; during hadra, 205 
Ghwal, characteristics of, 135-137 
Gifts (ziyara), to poor, 118-119 
Gnawa: in Meknes, 79, HO, 204, 209; 
considered black, 86; as curers, 135, 
158-163; belief in named jnun, 
141, 142; relation to Lalla Malika, 
147; as exorcist seer, 147, 226; and 
e ATsha Qandisha, 122 
Gonorrhea, causation, 114 
Grave plots, in zavviyas, 7 9- 80 
Grotto of e Aisha Qandisha, 65, 170. 
171,175,182 

Group support in Hamadsha ther¬ 
apy, 215=216 

Guilt, Moroccan feelings of, 227-228 
Guwwala, xi 

Gwal (drum), 35] played by foqra, 
83; in ceremonies, 122 

Habus, 67, 68, 21 
Hadiyya, 152. See also Sacrifice 
Hadra: performance of, xii-xiv; audi¬ 
ence reactions to, xiii, 193, 215— 
216; financing of, 13, 188, 207, 209; 
and foqra, 83, 88; and devotees, 89; 
as profession, 107; and fear reac¬ 
tions, 153; as therapy, 164, 167, 
169; at saints’ tombs, 175; during 
pilgrimage, 175, 176-178; and sacri¬ 
fices, 176, 188-189; preparations 
for, 187-189; madina performance 
of, 187-210; invitations to, 188, 
210; litanies during, 190-192; posi¬ 
tion of women in, 189; dance pat¬ 
terns of, 192-193,194.195-204,211; 
hot part of, 192-207. 211; trance 
patterns of, 195-202. 211; music of, 
204-207; cold part, 207-209, 211; 
of e AIsha Qandisha, 207. 209; self- 
mutilation in, 208; gnawiyya, 209; 
shantytown performances, 210— 
211. See also Dance; Headslashing; 
Trance 

al-Hafyan, 2_L See also Sidi 1-Hafyan 
Haircutting, at saints* tombs, 83, 84 
Hairstyle, among Meknes Hamadsha, 
109 


Hajj, 169 

Hal: in Sufism, 16; legendary origin 
of, 34,43; as experienced by foqra, 
83. 85. 206; experienced by ghiy- 
yata, 88; and baraka, 167; experi¬ 
enced at saint’s tombs, 174; defini¬ 
tion of, 195; and somnambulistic 
state, 195n; dance patterns during, 
196-198; and cold part of hadra, 
208; and hypnotic trance, 222 
Hallucinations, 199,214 
Hamadsha: definition of, lj and cur¬ 
ing, 4-5. 6-7. 9, 85-87, 132, 133, 
135, 141, 167-168, 169. 220, 224; 
position in Moroccan society, 7, 
110-111, 112, 126-128; spiritual an¬ 
cestry of, 21; trance of, 43, 219; 
headslashing, 45; instruments of 
mutilation, 36-37; in saints’ villages, 
73-74. 76; zawiyas of, 79-80; initia¬ 
tory symptoms, 113, 125; musem 
of, 115-118; and e ATsha Qandisha, 
43, 44. 143-146; therapeutic the¬ 
ories of. 134. 149-166. 166-168. 
212-215, 221. 223: and exorcist- 
seers, 162; stereotypes of, 219; feel¬ 
ings of inadequacy of, 229 
Hamadsha brotherhoods: position of, 
xv, lj 3; organization of, 3-4. 114- 
115, 125-128; membership in, 26, 
27-29, 87-88; ceremonies of, 27, 
85, 97-98, 107, 185-211; and miz- 
war, 71, 108. in madina of Meknes, 
79-97; election of muqaddim, 80- 
82. devotees of, 89-96, 111-112, 
125-128; finances of. 99. 113, 118- 
125; in shantytowns, 101, 107-111 
Hamadsha saints, l, 21; biography of, 
22-24; legends of, 30, 32-47; tombs 
of, 61, 64-65. 65-66; pilgrimages to, 
169-178, 178-184; in cures, 22L 
See also Sidi ‘Ali ben Hamdush; 
Sidi Ahmed Dghughi 
Hammu Qiya, 143. 146 
Harmal, 146,129 
Hasalban, 148 
Hashish, 233-234 

Head: symbolic of penis, 53; endow¬ 
ment of baraka, 162 








































INDEX 


253 


Head-slashing: audience reaction to 
xiii; description of, xiii, xiv; in 
legends, 15; origin of, 45; as sym¬ 
bolic castration, 53, 55, 227; prepa¬ 
ration for, 84; required by e Aisha 
Qandisha, 143; during trance, 200- 
201. 211, 167; similarity to Balinese 
kriss dancing, 200 
Heddawa, 22 

Henna: and baraka, 120; and Lalla 
Mira, 148; as cure for aching bones, 
171; used at c Aisha Qandisha’s 
grotto, 176,182 

Hizb: 17; definition, 80; example of, 
190. recitation of, 190-191 
Holy fool (majdub), 125 
Homosexuality: passive, 51; and pas¬ 
sage of baraka, 52 
Horm,62 

Hospitality, during musem, 122-123 
Hufra. See Grotto 
Hypocalcemia, 234 
Hypoglycemia, 234 
Hyperventilation, 192. 233 
Hypnosis, 231.232. 235 
Hysterical reactions, 97,212 

Ighud, 44 
Illergh, 142 

Illness: relation to jnun, 96-97 (see 
also Jnun); distinction between 
physical and mental, 134; taxon¬ 
omy of jinn-caused, 152-157: femi¬ 
nine connotation of, 224. 226. See 
also Diagnostics 

Images: in therapeutic therapy, 6, 
213-214: of psychic state, 223; of 
experience, 226n 

Immitation of animals: during hadra, 
145: camels and pigs, 201 
Incense: jnun aversion to, 138. See 
also Benzoin; Jawi; Harmal 
Initiation: into shantytown teams, 
108-109. 113; of camels and pigs, 
201 

Insanity, explanation for, 1 50-1 51 
Institutional baraka. See Baraka 
Iron: jnun aversion to, 138; in trance 
ceremony, 201 


ben 'Isa al-Mokhtari, Sidi M’hamed: 
21. 78; mausoleum of, 79; perform¬ 
ance at tomb of, 22. See also 
'Isawa; Isawiyya 

Tisawa, xiv, 26, 78-79; in shanty¬ 
towns, 107; belief in named jnun, 
141, 142; in trance attack, 203; 
theriomimetic behavior of, 143 
Isawiyya, 21, 29; in Meknes, 78-80. 
82: hadra of, 202 

Jawi, 138, 145, 146. See also Benzoin 
al-Jazuli, 20 

Jealousy, among wulad siyyid, 24 
Jeanmaire, FL, 158 
Jebel Zerhoun. See Zerhoun 
Jidba: definition, 88, 195, 219; and 
women, 196; and 'Aisha Qandisha, 
196; and men, 200; in therapy ex¬ 
planations, 219, 222 
Jilala: in Meknes, 79; in shantytowns, 
107; belief in named jnun, 141.142; 
and Lalla Malika, 147: and Lalla 
Mira, 148; and symbiotic cures, 
158-163; as exorcist-seers, 162 
Jilaliyya, 20, 26; hadra similar to 
hadra gnawiyya, 202 
al-jilani, Moulay Abdelqadar, 18 
Jinn. See jnun 

Jinniyya, 24, 44; commands of, 227. 

See also 'Aisha Qandisha; Jnun 
Jma', 61 

Jnun: definition, 2, 135; in legendary 
cycle, 41; attacks by, 91, 140. 152- 
157: and taifa, 107; and explanation 
for pathology, 135-149: in etiologi¬ 
cal theories, 134, 135; description 
of, 135-142; organization of, 139; 
taboos required by, 143: male, 148- 
149; as logical construct, 151; as 
cause of fear, 153: effect of trans¬ 
gression against, 156. 157: identifi¬ 
cation of, 159; commands of, 163— 

165. 227: and symbiotic cures, 159— 

166. See also Named jnun 
al-Junayd, 12 

Karama. See Miracles 
Kasba, 64 










254 


NDEX 


Katib, 82 

Khalifa: definition of, 81; duties of, 
82 

Khzana, 82 

Kif, in trance induction, 233-234 
Koranic teachers, powers of, 179. 
See also Fuqaha 


Lalla ‘ATsha. See 'Aisha Qandisha 
Lalla £ Auda, 33^ barren legend, 42 
Lalla Malika: description of, 146- 
147; response to dances by Gnawa 
Jilala, 142 

Lalla Mimuna: description of, 148 
Lalla Mira: description of, 147; re¬ 
sponse to hadras of Hamadsha and 
Jilala, 148 

Legends: hagiographic, 10, 30-31; 
themes, 31; cycle of, 31-34; expli¬ 
cation of, 35-47; interpretation of, 
47-56; in Hamadsha therapeutic 
practice, 213 
LeTourneau, 85 

Letter of Charter: as power of miz- 
war over muqaddim, 108; in shan¬ 
tytown team financial structure, 


Hi 

Levi-Strauss, Claude: resolution of 
conflict theory, 222-223 
Lienhardt, Godfrey: images as ele¬ 
ments of Hamadsha therapy the¬ 
ory, 6,213 

Lila: zawiya ceremony, 97; hadra 
performed at, 185 
Lineage of wulad siyyid, 67-71 
Litanies, in ceremonies, 186, See also 
Dikr; Hizh; Wanasa 
Lodges (zawiyas): definition of, 4; 
in Meknes, 75-79; description of, 
79-83: grave plots in, 79-80: organ¬ 
ization of, 80-81; foqra in, 83-89; 
muhibbin in, 89-91; Boualem, 91- 
97: ceremonies of, 79, 97-100. See 
also Zawiyas 

Luckmann, Thomas, 5-6. 7, 132 


Madina, in Meknes, 75-79: See also 
Lodges; Zawiya 


Madrub, description of, 155. See also 
Jnun 

al-Maghrawi, 121 
Mahdi ibn Tumart, 18 
Maklu c (fear), 153 
Malhun (lyrical song), 121 
Mamluk, 44. See also Jnun 
Maqama ( hal ), 16 
Maqius, 154. See also Possession 
Maraboutism, 1. 2.18, 52 
Marja, 128 

Marriages: between descendants of 
Sidi *Ali and Sidi Ahmed, 66-67; 
to 'Ai’sha Qandisha, 145 
Maryah, 153 
Maskun, 153, 156 
Massage, 182 
Massus, 152 
Masu^, 153 
Matrush, 152 
Mausoleums. See Tombs 
Meknes: description of, 75-79; his¬ 
tory of, 76; French in, 76-77; pop¬ 
ulation changes, 77; stereotype, 78; 
influence of Moulay Ismail, 77-78; 
brotherhoods of, 78-79; zawiyas 
nf, 79-100 

Men: position in Arab society, 9, 
226; stereotype of, 9; position of 
in hadra, 189; inability to play 
role, 224 

ben Meshish, Moulay ‘Abdcslem, 22 
el Merhasiyne, 63 

Miliana, xiv, 29; in shantytowns, 106- 
7; in Meknes, 22 

Mint: in cures, 179: as taboo, 183 
Miracles: in Sufism, 17; prophetic 
(mu'jizat), 17; frogs, 34; as proof 
(burhan), 30, 36 

Mizwar: as leaders, 63; as heads of 
saintly lineage, 67-68, 70; cheating, 
69: authority and function of, 71- 
72; succession of, 72; role in Ha¬ 
madsha zawiyas, 80-81; in shanty¬ 
town teams, 113, 115; at musem, 
116-118; 122-123; conflict with 
ideal behavior, 123-124 
Monkeys, in legend, 40 





















INDEX 


255 


Morality, and commands of jinniyya, 

222 

Mosques, in madinas, 10 6 
Moulay Brahim: pilgrimage to, 121 
Moulayldriss (saint): 32} in legend, 
37, 43; location of tomb, 62, 177; 
in dreams, 84 

Moulay Idriss (town), administra¬ 
tion of, 63 

Moulay Ismail, 18, 76, ^2, ^ 34, 35, 
36; conflict with Sidi c Ah, 40} at 
tomb of Sidi c Ali, 40^ influence on 
Meknes, 28 
Moulay Rachid, 31 
Msannad, 83, 84-85, 173, 128 
Muhibbin: definition of, 89} and za- 
wiyas, 89-91; characteristics of, 
96-97; in shantytown teams, 124— 
125, 127; and ‘Aisha Qandisha, 125 
Mulud, 111 

Muqaddim (Hamadsha): definition 
of, xii; role in ceremonies, xii, 161— 
162, 187, 188, 190, 207, 235} qualifi¬ 
cations for, 80} election of, 80-82; 
duties of, 82-83, 108; in talfa, 108— 
109; and devotees, 111-125; at 
saintly villages, 173,124 
Muqaddim (municipal), 104 
Musem, 45} in Zerhoun, 63, 115; 
property distribution at, 70; de¬ 
scription of, 115-118; sacrifices at, 
116; function of, 117-118; baraka 
at, 11; as faith renewal, 219 
Mushar, 154 
Mutism, 4 
Muttakil, 157 
Muwali, 112 

Nadir, 68 
Nafa c , 112 

Name-day celebrations, 74, 98-99 
Named-jnun: description of, 140, 
141-149; origin of belief in, 141; 
relationship to human beings, 142; 
in cures, 156-157, 162. See also 
Jnun; Symbiotic cures 
Naturalistic explanations, of illness, 
134 

Nira, 89 


Oboe. See Ghita 
Oboe players. See Ghiyyata 
Oil, in legends, 40-41 
Olive trees, 61, 62 

Ortigues, Marie-Cecile and Edmond, 

221 

Paques, Viviana, 141n 
Paraesthesia, 5,97, 199-200 
Paralysis: cured by Hamadsha, 4} 
resulting from attack by jnun, 113; 
classification of, 152-157 
Participational Mode of Responsibil¬ 
ity: defined, 151; role of jnun in, 
151-152; in explanation of therapy, 
213 

Patient: symptoms of, 4-5; participa¬ 
tion in Hamadsha therapy, 216- 
217; passivity of, 217; relationship 
with curer, 217-218 
Pathology: Hamadsha theory of, 134. 

135-149. See also Jnun 
Patrilineality, 7} and passage of bar- 
aka, 48, 52, 55, 56} and institutional 
baraka, 120-121; and wulad siyyid, 
67-71; importance in Moroccan so¬ 
cial organization, 224 
Patripotcstality, 224 
Phylacteries, 140. See also Amulets 
Pigs, imitation in trance, 198, 201 
Pilgrimage (ziyara): activities dur¬ 
ing, 2, 171-173, 176-178; made by 
devotees, 96} and musem, 115-118; 
and cures, 134n, 159. 169; cost of, 
159, 169; comparison with hajj, 
169; sequence of, 173-176; Fa¬ 
tima’s, 178-183. See also Hajj; Mu¬ 
sem 

Poisoning: in legends, 39} in thera¬ 
peutic theory, 149-150 
Pomegranate, 38 

Possession: Hamadsha cures of, 4, 
166, 231; classification of states of, 
151-157; and trance, 152-153n, 220} 
and exorcism, 157-159; in ancient 
world, 158; and symbiotic cures, 
159-166 

Prickly pears, in legends, 39 
Prostitutes, 62,121 























256 


INDEX 


Psychoanalysis, and Hamadsha cures, 
222-223 

Purification: and pilgrimages, 173; 
and hadra, 188 

Qadiriyya, IK 

Qariwiyyin University, 22, IS 
Qedesha, 143 
Qelaa, 63 

Qubba, 2. See also Tombs 
Race, 86 

Rahaliyya, 209. See also Sidi Rahal 
Ralimaniyya, 231. 213 
Rih: definition, 95] in Boualem’s net¬ 
work, 95] as command of jinn, 143; 
‘Aisha Qandisha’s, 145; in trance 
induction, 157; during hadra, 204. 
See also Trance; Hadra 
Rorschach responses, of hadra per¬ 
formers, 207 
Rue, 122 

Sacrifice: in therapeutic theory, 64; 
types of, 77; in ceremonies, 98, 176. 
188; to 'ATsha Qandisha, 98; at 
musem, 116; to saint, 174, 178; at 
'ar, 176; promise of by patient, 226. 
See also e Ar; Debiha; Dhiya; Gifts 
Sadaqa: definition of, 31* 185; name 
for Hamadsha ceremony, 185 
Saint: in Morocco, 1-3; and baraka, 
2-31; in Sufism, 16-17, 19; legends 
of, 30-31; in Beni Rachid, 65; in 
Meknes, 78-19; relations with, 81; 
in shantytowns, 105, 106, 107-108 ; 
pilgrimages to, 106, 169-170, 181; 
in cures, 134, 166-167; patient iden¬ 
tification with, 226, See also Sidi 
c Ali ben Hamdush; Sidi Ahmed 
Dghughi; and other individual en¬ 
tries 

Salwat al-Anfas, 21 
Sand, 120 
Santal, 142 

Sargent, William, 220 
Self-mutilation, 208-209. 222. See also 
Headslashing 


Semen: and baraka, 49-50, 226; origin 
of, 228n 

Sensory deprivation, 231-232 
Sensory overloading, 231-232 
Sexual intercourse: and passage of 
baraka, 121; practice of, 198, 228; 
after hadra, 2111 

al-Shadhili, 22, 80; and hizb of £ Alla- 
liyyin, 190 

Shamanistic cures, 222-223 
Shantytown: 7, 8; description of, 
101-103; and urbanization, 103-107 
Shantytown teams. See Taifa 
Sharif, 4,18,108 

Sharqi, Bu'abid, 21, 32, 35, 36, 37, 49, 
53, 228 
bel Shaqur, 64 
Shayatin, 135, 136-137 
al-Shayyaf, 20 
Sheikh, 2.17 

Sheikh al-Kamal, xi, xiv, 97, 128. See 
also ben c Isa; e Isawa; 'Isawiyya 
Sidi Ahmed Dghughi, xv, 2, 3, 21; 
biography of, 23-24; legends of, 
32—47 passim; and headslashing, 
47; and baraka, 48-56; adepts’ rela¬ 
tionship to, 84-85; and teams, 107- 
108; musem of, 115-118; and ex¬ 
planation of cures, 166; pilgrimage 
to, 169, 170, 171-178 passim , 181- 
182; symbolic associations of, 225— 
226. See also Saints; Wulad Siyyid 
Sidi Ahmed ou Mousa, 142 
Sidi c Ali ben Hamdush, xv, 2, 3, 21; 
biography of, 22-23. 24; brother¬ 
hoods of, 24-26; legends of, 32-47 
passim; and baraka, 48-56; tomb of, 
61, 64-65; descendants of, 66-74; 
adepts’ relationship to, 84-85; and 
teams, 107-108; and finances, 113; 
musem of, 115-118; and explana¬ 
tion of cures, 166; pilgrimage to, 
169, 170, 171-178 passim , 179-183; 
symbolic associations of, 225-226. 
See also Saints; Wulad Siyyid 
Sidi £ Ali Sanhaji, 21 
Sidi Baba, 102, 105 

Sidi 1-Hafyan, 64. See also al-Hafyan 





















INDEX 


257 


Sidi Hammu, 146 
Sidi 1-Hasan 1-Yussi, 4Q 
Sidi Heddi, 120 
Sidi 1-Husayn, 37 
Sidi Mejma', 116 
Sidi JVlimun, 148 
Sidi Musa (jinn), 148 
Sidi Musa (saint), 65 
Sidi Qasim, 38, 32 

Sidi Rahal: and origin of axe, 40; 

pilgrimage to, 121 
Sidi Shamharush, 44,142 
Sidina Bilal, 86 
Sidiqiyyin, 24 
Silsila, 17,18,52 
Silver, jnun’s aversion to, 138 
Slapping, 152 
Snakes, in legend, 32 
Somnambulistic, state, 125 
Spiny cactus, 40 

Spitting: in legend, 38; at saint’s tomb, 
84; in Madina ceremonies, 182 
Submission, theme in legends, 48 
ben Suda, 38 

Sufism: definition, 15; history of, 16- 
21; in Morocco, 18-21 
Suggestion, in therapy, 218. 20Q 
Symbiotic cures, 6, 158-166 
Symptomatology: of Hamadsha pa¬ 
tients, 5, 95, 113; relation in jnun, 
96-97; in choice of therapy, 134; 
after hadra, 210 
Syphilis, 144 

Tabbal, xi, 108 
Tabil, in hadra, 122 
Tabi c : definition of, 83; relation to 
foqra, 84 

Taboos, post pilgrimage, 183 
TaTfa: and zawiya, 4; in saintly vil¬ 
lages, 73-74; organization of, 101, 
107-111, 114-115; and baraka, 107, 
121, 122; stereotypes of, lH, 187; 
finances of, 113-118, 118-125 

passim; attitude of mizwar, 124; 
and devotees, 125-128; and cere¬ 
monies, 186, 187,210-211 
Talla e . See Exorcist-seer 


Tantrums, in women, 103n 
Tar: jnun aversion to, 138; and 
c ATsha Qandisha, 146 
Tariqa: definition of, 2, 4; practices 
of, 12 See also Brotherhoods 
Tasbih (beads), 45 
ben-Tashfin, Yusset, 26 
Teams. See Ta'ifa 
at-Tebba, 21 

Tension, between men and women, 

122 

Therapy, explanations of: symbolic 
interpretations, 6; women, 213; 
male-female symbolism, 213; im¬ 
ages, 213-214; anxiety reaction, 
214; group cathartic function, 215— 
216; illness, 215; patient participa¬ 
tion, 216: patient-curer relation¬ 
ship, 217; insight and suggestion, 
218; trance, abreaction, catharsis, 
219-230 

Therapy, theory of: definition, 5-6. 
11, 213; cost of, 134; and jnun, 135- 
149; and diagnostics, 149-157; ther¬ 
apeutic, 157-159; and symbiotic 
cures, 159-168; and saints, 170; 
situation centered, 184 
Tijaniyya, 20. 79.80 
Tombs: Hamadsha, description of, 
62, 64-65. 65-66. 181; proceeds 
from, 67; of Sidi ben Isa, 79; cere¬ 
monies at, 83-84; procedures 
within, 18L 182. See also Saints 
Touijer, Mohammed, 31* 81 
Trance: description of, xiii; in wulad 
siyyid, 74; and baraka, 121; and 
jnun, 156; in therapeutic practices, 
167; awakening from, 186: defini¬ 
tion, 195; comparison with Bali¬ 
nese, 196: during hadra, 195-202. 
204. 211; animal imitation during, 
201; emotional exhaustion and col¬ 
lapse, 220-221: suggestibility, 220- 
221. regressive behavior, 221-222; 
and hal, 222: and images, 223. See 
also Hal; Jidba 
Trance induction, 231-235 
Tranfer agents, 118-125 passivt 




















258 


NDEX 


Transference, 217, 235 
Transvestite, 196n 

Tuba: in legends, 44; use at shrines, 
175* 176; in cures, 182 
Tuberculosis, 134 
Tuhama, 29 

Urbanization: and madina of Mek- 
nes, 75-79; and shantytowns of 
Meknes, 103-107; and organization 
of devotees, 125 - 128 

Virility, and illness, 213 
Vodun spirits, 138, 142 
Volubilis, 35* 62 

Vomit, in legends, 32, 35, 36, 37,41 
Vows, to ‘Aisha Qandisha, 65. See 
also Ahd; c Ar 

Wali, L See also Saint 
Wanasa, 19L See also Litanies 
Wasp, 201 

Wealth: division among wulad siy- 
yid, 68* 69-71; division of zawiya, 
82, 99-100; division among teams, 
107, 113; and baraka exchange, 
118-125 

Westermarck, Edward, 135* 136, 
137-138, 141. 143 
Windus, John, 26 


Witchcraft, 149 

Women: social position of, 8-9. 98; 
stereotypes of, 8-9, 56, 213. 224- 
225; and baraka, 48-49; tantrums 
of, 103; and hadra, 112, 189, 192. 
196-199. 209; and jnun, 138* 142^ 
as exorcist-seers, 162; fear of, 213 
Wulad Khalifa, 26, 110,204 
Wulad Majdub, 25* 26 
Wulad siyyid: organization of, 4* 
68-71, 115: and mizwar, 71-72; and 
baraka, 73-74, 118-125; and trance, 
24 

Yabis, 199 

Zar spirits, 135>, 142 
Zawiya: definition of, 4; in Meknes, 
75* 79-83; role of mizwar in, 80; 
officers in, 80* 82-83; caretaker of, 
83; adepts of, 83-89; devotees of, 
89-97; ceremonies of, 97-99, 187— 
210; distribution of income, 99- 
100, 118-125; in Hamadsha com¬ 
plex, 114-115; in musem, 117; in 
exchange system, 118-125; com¬ 
pared with shantytown teams, 125— 
128. See also Brotherhood 
Zerhoun, 61-63 

Ziyara, 169, See also Pilgrimage 





















Copyrighted material 








University of California Press, Berkeley 94720